ext/celex/62023TJ0328
JUDGMENT OF THE GENERAL COURT (Ninth Chamber)
4 February 2026 ( * )
( Research and technological development – Horizon Europe Framework Programme for Research and Innovation (2021-2027) – High-performance computing applications – Call for proposals with a view to the award of a grant – Decision rejecting a proposal – Decision rejecting the request for an evaluation review – Action for annulment – Locus standi – Interest in bringing proceedings – Admissibility – Manifest error of assessment – Obligation to state reasons – Principle of transparency – Equal treatment – Article 30 of Regulation (EU) 2021/695 )
In Case T‑328/23,
Radical-Consulting UG, established in Randowtal (Germany),
Surf BV, established in Amsterdam (Netherlands),
Bayerische Akademie der Wissenschaften, established in Munich (Germany), represented by M. Martens, K. Munungu and B. Mourisse, lawyers,
applicants,
v
European High Performance Computing Joint Undertaking (EuroHPC Joint Undertaking), represented by P.-E. Partsch, B. ten Seldam and S. Thomas, lawyers,
defendant,
THE GENERAL COURT (Ninth Chamber),
composed, at the time of the deliberations, of L. Truchot, President, M. Sampol Pucurull (Rapporteur) and T. Perišin, Judges,
Registrar: V. Di Bucci,
having regard to the written part of the procedure,
having regard to the order of 23 February 2024, by which it reserved its decision on the plea of inadmissibility raised by the EuroHPC Joint Undertaking until it rules on the substance of the case,
having regard to the measure of organisation of procedure of 3 March 2025 and the response of the applicants lodged at the Court Registry on 19 March 2025,
having regard to the fact that no request for a hearing was submitted by the parties within three weeks after service of notification of the close of the written part of the procedure, and having decided to rule on the action without an oral part of the procedure, pursuant to Article 106(3) of the Rules of Procedure of the General Court,
gives the following
Judgment
1 By their action under Article 263 TFEU, the applicants, Radical-Consulting UG, Surf BV and Bayerische Akademie der Wissenschaften, seek the annulment of, first, Decision Ares(2022) 5697480 of the European High Performance Computing Joint Undertaking (EuroHPC Joint Undertaking) of 11 August 2022 (‘the contested decision’), by which it rejected the grant proposal for the CompBioMed3 project with reference 101092620, submitted by the consortium of which they are members (‘the consortium’) in call for proposals HORIZON‑EUROHPC‑JU‑2021‑COE‑01 under the Horizon Europe Framework Programme for Research and Innovation (‘the Horizon Europe programme’), and, second, Decision Ares(2023) 2473635 of the EuroHPC Joint Undertaking of 4 April 2023, by which it rejected the request for an evaluation review of the grant proposal.
I. Background to the dispute
2 The Horizon Europe programme was established by Regulation (EU) 2021/695 of the European Parliament and of the Council of 28 April 2021 establishing Horizon Europe – the Framework Programme for Research and Innovation, laying down its rules for participation and dissemination, and repealing Regulations (EU) No 1290/2013 and (EU) No 1291/2013 (OJ 2021 L 170, p. 1, ‘the Horizon Europe Regulation’).
3 Council Decision (EU) 2021/764 of 10 May 2021 established the Specific Programme implementing Horizon Europe – the Framework Programme for Research and Innovation, and repealed Decision 2013/743/EU (OJ 2021 L 167 I, p. 1).
4 The evaluation of proposals in calls for proposals under the Horizon Europe programme, is carried out, in accordance with Article 28(1) of the Horizon Europe Regulation, on the basis of the following three award criteria: ‘excellence’, ‘impact’ and ‘quality and efficiency of the implementation’. Under Article 28(3) of that regulation, the further details concerning the application of those award criteria and their weighting and thresholds are to be set out in a work programme.
5 The European Commission adopted Implementing Decision C(2021) 1940 of 31 March 2021 on the adoption of the work programme for 2021-2022 within the framework of the Specific Programme implementing Horizon Europe – the Framework Programme for Research and Innovation and on its financing (‘the work programme’). Part D of the general annexes to the work programme defines the ‘excellence’, ‘impact’ and ‘quality and efficiency of the implementation’ award criteria for proposals in the context of different actions under the Horizon Europe programme, including research and innovation actions.
6 It was stated, under the heading ‘Scores and weighting’ in Part D of the general annexes to the work programme, that ‘each criterion [would] be scored out of 5’, that ‘the threshold for individual criteria [would] be 3’ and that ‘the overall threshold, applying to the sum of the three individual scores, [would] be 10’. It was also specified that ‘proposals that pass[ed] the individual threshold and the overall threshold [would] be considered for funding, within the limits of the available call [for proposals] budget’ and that ‘other proposals [would] be rejected’.
7 The EuroHPC Joint Undertaking is a joint undertaking established between the European Union, Member States and associated countries, as well as private partners, by Council Regulation (EU) 2021/1173 of 13 July 2021 on establishing the European High Performance Computing Joint Undertaking and repealing Regulation (EU) 2018/1488 (OJ 2021 L 256, p. 3, ‘the EuroHPC Regulation’).
8 According to Article 3(1) of the EuroHPC Regulation, the mission of the EuroHPC Joint Undertaking is inter alia to develop leading supercomputing infrastructures (acquisition of supercomputers), to develop applications optimised for those systems (in particular by supporting centres of excellence) and to provide access to supercomputing resources and services (through the allocation of the Union’s access time to the EuroHPC Joint Undertaking’s supercomputers). That mission is implemented according to the pillars of activity of the EuroHPC Joint Undertaking specified in Article 4 of that regulation.
9 Under Article 1(b) of the Statutes of the EuroHPC Joint Undertaking, which are annexed to the EuroHPC Regulation, the EuroHPC Joint Undertaking is to launch calls for proposals, calls for tenders, and any other instrument or procedure provided for in the Horizon Europe programme, with a view to creating a digital Europe and a facility for European interconnection.
10 On 15 December 2021, the Governing Board of the EuroHPC Joint Undertaking adopted, on the basis of the EuroHPC Regulation, Decision No 28/2021 adopting the Joint Undertaking’s work plan for the year 2022 (‘the 2022 work plan’). The 2022 work plan defines the actions to be implemented in 2022, including calls for proposals to be launched in that year, with a view to the creation of centres of excellence for high-performance computing applications. The 2022 work plan defines the specific conditions applying to those calls for proposals.
11 On 27 January 2022, the EuroHPC Joint Undertaking published call for proposals EUROHPC‑JU‑2021‑COE‑01, entitled ‘Centres of Excellence preparing applications in the Exascale era’, on the Commission’s website (‘the call for proposals’). The call for proposals comes under ‘research and innovation actions’. The goal of the call for proposals is, in essence, to develop or rewrite the codes of certain applications towards exascale performance mainly for scientific challenges. Exascale computing refers to computing systems capable of performing ten to the power of 18 (10 18 ) simultaneous computations per second. The call for proposals provides inter alia that the award criteria, scoring and thresholds are laid down in Part D of the general annexes to the work programme. The specific conditions for the call are described in the 2022 work plan.
12 On 6 April 2022, the consortium, which is composed of 11 members, including the applicants, submitted a grant proposal for CompBioMed3 (with reference 101092620) in the field of computational biomedicine at the exascale (‘the proposal’). The objective of that proposal was to develop the third phase of the CompBioMed Centre of Excellence, which is a centre of excellence funded by the Horizon 2020 programme, aimed at the use and development of computational methods for biomedical applications. The University of Amsterdam performed the role of coordinator for the call for proposals (‘the coordinator’). As such, it sent and received all communications on behalf of the consortium in the context of that call for proposals.
13 By the contested decision, the EuroHPC Joint Undertaking informed the consortium that the proposal had not been selected to receive the requested funding on the ground that it did not reach the minimum threshold necessary. The EuroHPC Joint Undertaking enclosed with the contested decision an evaluation summary report (‘the evaluation report’), which showed the evaluators’ comments and the scores awarded. According to the evaluation report, the proposal was awarded scores of 2.5 out of 5 for the ‘excellence’ criterion, 3 out of 5 for the ‘impact’ criterion and 2.5 out of 5 for the ‘quality and efficiency of the implementation’ criterion and, therefore, a total score of 8 out of 15.
14 On 28 July and 31 August 2022, the consortium submitted an application for access to documents pursuant to Regulation (EC) No 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents (OJ 2001 L 145, p. 43). The application covered all documents held by the EuroHPC Joint Undertaking relating to the call for proposals.
15 On 7 September 2022, the EuroHPC Joint Undertaking refused the application for access to documents based on the exceptions laid down in the second subparagraph of Article 4(3) of Regulation No 1049/2001 concerning protection of the decision-making process and in the first indent of Article 4(2) of that regulation concerning protection of commercial interests.
16 On 9 September 2022, the coordinator submitted to the EuroHPC Joint Undertaking a request for review of the evaluation report in accordance with the procedure laid down in Article 30 of the Horizon Europe Regulation, claiming factual errors, manifest errors of assessment and infringement of the obligation to state reasons and the principle of transparency enshrined in Article 29(2) of that regulation.
17 On 28 September 2022, the consortium made a confirmatory application for access to documents under Article 7(2) of Regulation No 1049/2001.
18 By letter of 18 October 2022, the EuroHPC Joint Undertaking granted full access to Annex 4, entitled ‘Topic-specific statistics’, to the evaluation report for the call for proposals, only partial access to the evaluation summary reports for the other proposals submitted in that call and to other documents relating to the evaluation report for the call for proposals (evaluation report for the call for proposals, Annex 1 ‘call ranked list’, Annex 2 ‘list of below-threshold proposals’, Annex 3 ‘country-specific statistics’ and Annex 5 ‘minutes of the deliberations of the assessment committee’) and denied access to the proposals submitted. The EuroHPC Joint Undertaking justified the partial access and the denial of access on the basis of the exceptions laid down in the first indent of Article 4(2) and Article 4(1)(b) of Regulation No 1049/2001.
19 By email of 22 December 2022, the consortium submitted an amended request for review after consulting the documents sent by the EuroHPC Joint Undertaking further to its application for access, by which it also claimed infringement of the principle of equal treatment.
20 By letter of 4 April 2023, received the following day by the coordinator, the EuroHPC Joint Undertaking rejected the consortium’s request for review (‘the decision rejecting the request for review’).
II. Forms of order sought
21 The applicants claim that the Court should:
– annul the contested decision;
– annul the decision rejecting the request for review;
– order the EuroHPC Joint Undertaking to pay the costs.
22 The EuroHPC Joint Undertaking contends, in essence, that the Court should:
– dismiss the action as manifestly inadmissible;
– in the alternative, dismiss the action as unfounded;
– order the applicants jointly and severally to bear the costs.
III. Law
A. Admissibility of the claim for annulment
23 The EuroHPC Joint Undertaking raises three pleas of inadmissibility concerning, first, the absence of locus standi of the applicants, second, the absence of their interest in bringing proceedings and, third, the principle of ‘ nul ne plaide par procureur ’ – one cannot sue or be sued by proxy.
1. Locus standi
24 The EuroHPC Joint Undertaking claims that the contested decision and the decision rejecting the request for review (together, ‘the contested decisions’) are not addressed to the applicants for the purposes of the fourth paragraph of Article 263 TFEU. They were addressed to the coordinator alone. The applicants must therefore show that the contested decisions are of direct and individual concern to them, which they have not established.
25 In the first place, the EuroHPC Joint Undertaking submits that, according to the conditions laid down in the call for proposals, the contested decisions are of concern only to the consortium, and not its individual members. The EuroHPC Joint Undertaking observes in that regard that, pursuant to Article 22(2) of the Horizon Europe Regulation and the general annexes to the work programme, the call for proposals was open only to consortia of at least three entities established in three different Member States or associated countries. Therefore, only the consortium could be directly and individually concerned.
26 For a consortium to be able to apply for the annulment of a decision concerning it, the EuroHPC Joint Undertaking maintains that it should be validly represented by its coordinator or a majority of its members. In that regard, the EuroHPC Joint Undertaking observes that the coordinator is not included among the applicants and that two thirds of the members of the consortium have not challenged the contested decisions. Furthermore, the applicants originate from only two Member States, Germany and the Netherlands, and do not therefore meet the requirements governing the call for proposals.
27 In the second place, the EuroHPC Joint Undertaking claims that the absence of direct and individual concern for the applicants by the contested decisions is demonstrated by the facts.
28 In that respect, the EuroHPC Joint Undertaking observes that the consortium acted individually, through the coordinator, in every interaction related to the call for proposals leading up to the present application, including the request for review of the evaluation report mentioned in paragraph 16 above. The EuroHPC Joint Undertaking notes that the contested decisions were addressed to the coordinator and asserts that it never interacted directly with the applicants in connection with the call for proposals.
29 Thus, according to the EuroHPC Joint Undertaking, the proposal is a collective offer from the whole consortium, which pooled the individual skills of each member for the purposes of a common project. The EuroHPC Joint Undertaking asserts that the application shows an abrupt and total disengagement of the project coordinator. There is no indication that the consortium desired to act for the annulment of the contested decisions.
30 In the third place, the EuroHPC Joint Undertaking submits that if grievances from single members of a large consortium were admissible, that would significantly impact the proper administration of justice. Furthermore, grievances raised by an insignificant minority of the members of a large consortium would disproportionately disrupt the functioning of the EuroHPC Joint Undertaking.
31 The applicants dispute the EuroHPC Joint Undertaking’s arguments.
32 Under the fourth paragraph of Article 263 TFEU, ‘any natural or legal person may … institute proceedings against an act addressed to that person or which is of direct and individual concern to them …’.
33 In the present case, it should be noted that the application was lodged by three of the 11 members making up the consortium. The coordinator is not included among the applicants.
34 Furthermore, the applicants claim that they are not acting as representatives of the consortium, but as members of it.
35 While the contested decisions are, from a formal point of view, addressed to the consortium through the coordinator, the fact remains that, as is evident from the written submissions of the applicants, without that being disputed by the EuroHPC Joint Undertaking, the consortium has never had legal personality. Accordingly, pursuant to Article 263 TFEU, since that ad hoc structure was transparent so far as its members are concerned, the members of the consortium must all be regarded as addressees of the contested decisions. It was therefore open to the applicants, as addressees of the contested decisions, to challenge them under the conditions laid down by Article 263 TFEU (see, to that effect, judgment of 23 May 2014, European Dynamics Luxembourg v ECB , T‑553/11, not published, EU:T:2014:275, paragraph 84 and the case-law cited). Consequently, there is no need to determine whether the contested decisions are of direct and individual concern to the applicants within the meaning of the fourth paragraph of Article 263 TFEU, contrary to the claim made by the EuroHPC Joint Undertaking.
2. Interest in bringing proceedings
36 The EuroHPC Joint Undertaking disputes the applicants’ interest in bringing proceedings since, in the first place, any annulment of the contested decisions would not have an erga omnes effect and they would therefore remain in force in respect of the other eight members of the consortium which did not challenge them. The EuroHPC Joint Undertaking relies in that regard on the case-law arising from the judgment of 14 September 1999, Commission v AssiDomän Kraft Products and Others (C‑310/97 P, EU:C:1999:407).
37 Furthermore, the EuroHPC Joint Undertaking observes that calls for proposals launched by it concern highly complex supercomputing projects, which require the involvement of large consortia made up of numerous individual members.
38 According to the EuroHPC Joint Undertaking, an erga omnes annulment of one of its decisions at the request of an insignificant minority of the members of any consortium would significantly affect both the proper administration of justice and the principle of legal certainty and would be contrary to the requirements of procedural economy. EU bodies would be exposed to litigation even if the majority of the consortium agreed with the decision at issue or at least did not oppose it.
39 The EuroHPC Joint Undertaking also asserts that the principle of legal certainty precludes the obligation to re-examine the contested decisions in respect of the other eight members of the consortium or the consortium itself.
40 In the second place, the EuroHPC Joint Undertaking claims that it is unlikely that the applicants will procure an advantage from any annulment of the contested decisions. The EuroHPC Joint Undertaking asserts that such annulment would not change the position of the consortium or of its other eight members. If the contested decisions were annulled specifically for the applicants, they would not be able to produce the work in the proposal on their own.
41 First, the EuroHPC Joint Undertaking notes that the applicants do not fulfil the eligibility criteria set out in the call for proposals, which are laid down for consortiums on the basis of the Horizon Europe Regulation (see paragraphs 25 and 26 above). The EuroHPC Joint Undertaking also maintains that the applicants do not dispute those criteria in the present action. Second, the EuroHPC Joint Undertaking claims that the applicants have not demonstrated that they could, on their own, satisfy the technical conditions laid down in the call for proposals.
42 The applicants dispute the EuroHPC Joint Undertaking’s arguments.
43 It should be noted that, according to settled case-law, an action for annulment brought by a natural or legal person is admissible only in so far as that person has an interest in the annulment of the contested measure, which presupposes that annulment of that measure must of itself be capable of having legal consequences and that the action must be likely, if successful, to procure an advantage for the party who brought it (see judgment of 18 January 2018, Kenup Foundation and Others v EIT , T‑76/15, EU:T:2018:9, paragraph 27 (not published) and the case-law cited).
44 As regards the first complaint concerning the limited effects of any annulment of the contested decisions, it should be noted that the annulment of a decision rejecting a proposal submitted in a call in the field of research and technological development has an erga omnes effect, which could benefit all the members of the consortium (see, to that effect, judgment of 18 January 2018, Kenup Foundation and Others v EIT , T‑76/15, EU:T:2018:9, paragraph 32 (not published)), and not only those which brought the action.
45 The EuroHPC Joint Undertaking’s arguments concerning the judgment of 14 September 1999, Commission v AssiDomän Kraft Products and Others (C‑310/97 P, EU:C:1999:407), are not such as to call that finding into question. As the applicants point out, that judgment concerned the effects of a decision adopted by the Commission on the basis of Article 101 TFEU. Although drafted and published as a single measure, this kind of decision must be regarded as a group of individual decisions establishing, in relation to each of the undertakings to which it is addressed, the breach or breaches which that undertaking has been found to have committed and, where appropriate, imposing on it a fine. In those circumstances, such decisions can be annulled only with respect to those addressees which have successfully brought an action before the EU judicature, and remain binding on those addressees which have not applied for their annulment (see judgment of 15 October 2002, Limburgse Vinyl Maatschappij and Others v Commission , C‑238/99 P, C‑244/99 P, C‑245/99 P, C‑247/99 P, C‑250/99 P to C‑252/99 P and C‑254/99 P, EU:C:2002:582, paragraph 100 and the case-law cited).
46 In the present case, as was stated in paragraphs 35 and 44 above, because the contested decisions are addressed to the consortium, which is a transparent ad hoc structure, their possible annulment would have an erga omnes effect and would therefore benefit all the members of the consortium.
47 The EuroHPC Joint Undertaking’s arguments that an erga omnes annulment at the request of an insignificant minority of the members of a consortium, if it were decided, would affect the proper administration of justice, the principle of legal certainty and the requirements of procedural economy must be rejected in so far as, if accepted, they would deprive those members of their right of individual action for the purposes of the fourth paragraph of Article 263 TFEU, as was recalled in paragraph 35 above.
48 The second complaint raised by the EuroHPC Joint Undertaking concerning the limited likelihood that the applicants will procure an advantage from any annulment of the contested decisions is based on the premiss that such annulment would not change the situation of the consortium or of its other members. However, as is stated in paragraph 46 above, such annulment would benefit all the members of the consortium.
49 The applicants had an interest in the proposal of the consortium to which they belong being selected and, consequently, in challenging the rejection of that proposal by the EuroHPC Joint Undertaking. Annulment of the contested decisions would, at least, give them an opportunity for that proposal to be selected. Failing that, that annulment could constitute the basis of an action for damages (see, to that effect, judgment of 18 January 2018, Kenup Foundation and Others v EIT , T‑76/15, EU:T:2018:9, paragraph 31 (not published) and the case-law cited).
50 It should also be noted that, according to case-law, recognition of the interest in bringing proceedings cannot be made dependent on the submission by the applicants of evidence that if the alleged illegalities and manifest errors of assessment had not taken place, they would have had a genuine chance of the proposal of the consortium to which they belong being selected and, consequently, a claim for compensation being justified (judgment of 18 January 2018, Kenup Foundation and Others v EIT , T‑76/15, EU:T:2018:9, paragraph 34 (not published)).
51 In those circumstances, the applicants must be considered to have an interest in applying for the annulment of the contested decisions.
52 It follows that the first two pleas of inadmissibility raised by the EuroHPC Joint Undertaking must be rejected.
53 Lastly, with regard to the arguments put forward in support of the third plea of inadmissibility, namely that the applicants could act neither on behalf of the majority of the members of the consortium nor on behalf of the coordinator, it is sufficient to recall that the applicants are not acting as representatives of the consortium, but as members of it.
54 It follows from the foregoing that the action is admissible.
B. The merits of the claim for annulment
55 The applicants are seeking the annulment of, first, the contested decision and, second, the decision rejecting the request for review.
1. The claim for annulment directed against the contested decision
56 In support of their application for annulment of the contested decision, the applicants rely on four pleas in law, alleging, first, in essence, a manifest error of assessment and infringement of the principle of good administration, second, infringement of the obligation to state reasons, third, infringement of the principle of transparency and, fourth, infringement of the principle of equal treatment. The Court considers it appropriate first to examine the second plea in law, then the third plea in law, followed by the first plea in law and, lastly, the fourth plea in law.
(a) The second plea in law, alleging infringement of the obligation to state reasons
57 The applicants submit that the EuroHPC Joint Undertaking infringed its obligation under the second paragraph of Article 296 TFEU in so far as the reasons set out in the evaluation report regarding the ‘excellence’ and ‘impact’ criteria do not provide them with sufficient information to know whether the contested decision may have been vitiated by an error enabling its validity to be challenged.
58 With regard to the ‘excellence’ criterion, the applicants put forward two complaints. First, they claim that the report does not explain the reasons why the baseline was not addressed sufficiently and convincingly. Second, they submit that the statement of reasons does not allow them to understand in a clear and precise manner the reasons why the detailed description of open science practices in their proposal is not adequate and does not explain the way in which the connection with data management should have been highlighted in the proposal.
59 With regard to the ‘impact’ criterion, the applicants put forward three complaints. They claim, first of all, that the evaluation report does not explain the reasons why the EuroHPC Joint Undertaking considered that the dissemination plan was generic and that the target groups were not adequately described. Second, in their view, the report also fails to state the reasons why the connection between selected applications and market needs was not clearly shown in the proposal. Lastly, the evaluation report does not explain the reasons why aspects relating to intellectual property rights, exploitation rights and confidentiality issues were not adequately addressed.
60 The EuroHPC Joint Undertaking disputes the applicants’ arguments.
61 According to settled case-law, the statement of reasons required under Article 296 TFEU must disclose in a clear and unequivocal fashion the reasoning followed by the institution which adopted the measure in such a way as to enable the person concerned to ascertain the reasons for the measure and to enable the court having jurisdiction to exercise its power of review (see judgment of 29 September 2011, Elf Aquitaine v Commission , C‑521/09 P, EU:C:2011:620, paragraph 147 and the case-law cited).
62 Thus, in the context of individual decisions, it is settled case-law that the purpose of the obligation to state the reasons on which such decisions are based is, in addition to permitting review by the Courts, to provide the person concerned with sufficient information to know whether the decision at issue may be vitiated by an error enabling its validity to be challenged (see judgment of 29 September 2011, Elf Aquitaine v Commission , C‑521/09 P, EU:C:2011:620, paragraph 148 and the case-law cited).
63 The statement of reasons required by Article 296 TFEU must be appropriate to the act at issue and the context in which it was adopted. It is settled case-law that the requirement to state reasons must be assessed by reference to the circumstances of the case, in particular the content of the measure in question, the nature of the reasons given and the interest which the addressees of the measure, or other parties to whom it is of direct and individual concern, may have in obtaining explanations. It is not necessary for the reasoning to go into all the relevant facts and points of law, since the question whether the statement of reasons meets the requirements of Article 296 TFEU must be assessed with regard not only to its wording but also to its context and to all the legal rules governing the matter in question (see judgments of 29 September 2011, Elf Aquitaine v Commission , C‑521/09 P, EU:C:2011:620, paragraph 150 and the case-law cited, and of 9 June 2021, Ryanair v Commission (Condor; Covid-19) , T‑665/20, EU:T:2021:344, paragraph 39 and the case-law cited).
64 In particular, the reasons given for a measure adversely affecting a person are sufficient if that measure was adopted in a context which was known to that person and which enables him or her to understand the scope of the measure concerning him or her (see judgment of 15 November 2012, Council v Bamba , C‑417/11 P, EU:C:2012:718, paragraph 54 and the case-law cited).
65 Lastly, it must be borne in mind that the obligation to state reasons established in Article 296 TFEU is an essential procedural requirement which must be distinguished from the question whether the reasoning is well founded, which is concerned with the substantive legality of the measure at issue. The reasoning of a decision consists in a formal statement of the grounds on which that decision is based. If those grounds are not supported or are vitiated by errors, such defects will vitiate the substantive legality of the decision, but not the statement of reasons in it (see judgment of 23 February 2022, United Parcel Service v Commission , T‑834/17, EU:T:2022:84, paragraph 188 and the case-law cited).
66 Furthermore, Article 29(4) of the Horizon Europe Regulation provides that ‘in accordance with Article 200(7) of the [applicable] Financial Regulation, applicants shall receive feedback at all stages of the evaluation and, where the proposal is rejected, the reasons for rejection’.
67 It should be noted at the outset that some of the applicants’ arguments raised in connection with the plea in law alleging infringement of the obligation to state reasons actually relate to the merits of the assessment in the evaluation report. Indeed, in support of their arguments concerning the ‘excellence’ criterion, the applicants claim in particular that the proposal discusses the baseline of the research and open science practices, contrary to what is stated in the EuroHPC Joint Undertaking’s report. With regard to the ‘impact’ criterion, the applicants assert that the proposed workplans targeted the expected outcomes and impacts. Those arguments thus relate to the question whether the reasoning is well founded and, in the light of the case-law recalled in paragraph 65 above, they are irrelevant in support of the present plea in law. Consequently, they will be examined in the context of the third plea in law alleging a manifest error of assessment.
68 In the present case, with regard to the evaluation of the ‘excellence’ criterion in the evaluation report, it should be noted that the EuroHPC Joint Undertaking assessed the objectives of the proposal, the project’s envisaged improvements, the project’s capacity to go beyond the state of the art, the proposed scientific methodology to reach exascale performance, artificial intelligence methods, the proposed co-design approaches, interdisciplinary approaches, open science practices and the gender dimension. For each of those subjects, the EuroHPC Joint Undertaking stated whether the proposal addressed convincingly the way in which those aspects would be dealt with in the project or provided sufficient details. The EuroHPC Joint Undertaking highlighted the weaknesses of the proposal. In particular, it stated that the proposal did not convincingly justify the need for exascale performance in order to implement the project, which was a major weakness.
69 It should be observed that the EuroHPC Joint Undertaking did explain, concisely but in a clear and unequivocal fashion, the reasons underlying the score awarded for the ‘excellence’ criterion. The evaluation report therefore discloses explicitly and sufficiently the reasons why the proposal obtained a score below the required threshold for that criterion, namely a score of 2.5 points out of 5.
70 With regard to the ‘impact’ criterion, in the evaluation report the EuroHPC Joint Undertaking assessed the project’s impact in the target groups mentioned, noting the insufficient impact of the project in the medical area. It also examined the dissemination plan and exploitation measures, once again considering that their connection with the medical area was insufficient. The EuroHPC Joint Undertaking thus stated that the dissemination plan was ‘generic’ and did not adequately describe the target groups and the potential users of applications developed in the project. It took the view that the project’s activities and interactions were, in fact, targeted at the academic and scientific sector. The EuroHPC Joint Undertaking also assessed potential barriers to the project and framework conditions which could have an influence on the impact of the project, considering them to be reasonable. Lastly, it stated that intellectual property rights, exploitation rights and confidentiality issues were not adequately addressed in the consortium’s proposal.
71 The assessments mentioned in paragraph 70 above, which are set out in the evaluation report and are accompanied by clarifications regarding the extent of the deficiencies identified, make it possible, on the one hand, to link them to the requirements laid down in the call for proposals and, on the other hand, to understand why the proposal did not obtain more than 3 points out of 5 for the ‘impact’ criterion.
72 The applicants’ complaints, as set out in paragraphs 58 and 59 above, are not such as to call that finding into question. The EuroHPC Joint Undertaking is not obliged to set out the facts and the legal considerations having decisive importance in the context of its decision (see, to that effect, judgment of 8 July 2009, Zenab v Commission , T‑33/06, not published, EU:T:2009:250, paragraph 143 and the case-law cited).
73 It should also be stated that the alleged failure to state reasons did not prevent the applicants from making arguments, as was noted in paragraph 67 above. The applicants were, in particular, in a position effectively to challenge the merits of the assessment of the description of the baseline. They were also able to challenge the assessment of open science practices and to claim that the proposed workplans targeted the expected outcomes and impacts.
74 It must therefore be held that the evaluation report accompanying the contested decision explains, in a clear and unequivocal fashion, the reasons having decisive importance in the context of that decision for which the EuroHPC Joint Undertaking rejected the proposal.
75 In those circumstances, the present plea in law must be rejected.
(b) The third plea in law, alleging infringement of the principle of transparency
76 The applicants claim that the EuroHPC Joint Undertaking applied award criteria which were not provided for either in the call for proposals or in any other document published as part of that call for proposals, which constitutes an infringement of the principle of transparency enshrined, inter alia, in Article 29(3) of the Horizon Europe Regulation. The applicants maintain that if they had known of those criteria, they would have focused more on those aspects and obtained additional points, which would have allowed them to meet all the required thresholds.
77 In the first place, with regard to the ‘excellence’ criterion, the applicants submit that the call for proposals did not require open science practices to be addressed in relation to data management.
78 In the second place, with regard to the ‘impact’ criterion, the applicants submit, first, that the call for proposals did not require the impact of their project on the targeted clinical areas to be identified, but the development of exascale to be demonstrated.
79 Second, the applicants claim that, by stating that the connection between selected applications and community or market needs is not set out, at least not clearly enough, in the proposal made by the consortium to which they belong, the EuroHPC Joint Undertaking departed from the envisaged award criterion according to which proposals had to show that the proposed workplans would be able to achieve the outcomes and impacts identified. The applicants add that the description in the call for proposals, which is cited by the EuroHPC Joint Undertaking in the defence, is not an award criterion.
80 Third, the applicants claim that neither the call for proposals nor the documents in that call for proposals which are cited by the EuroHPC Joint Undertaking in the defence require them to conclude a consortium agreement before submitting their proposal or to include within the proposal an intellectual property rights or exploitation rights plan or even to regulate those issues in detail at the proposal submission stage.
81 In the third place, with regard to the ‘quality and efficiency of the implementation’ criterion, the applicants claim that neither the call for proposals nor the other documents published by the EuroHPC Joint Undertaking as part of that call for proposals require the direct and sufficient involvement of health professionals in the development of high-performance computing applications in a centre of excellence with a medical focus.
82 The EuroHPC Joint Undertaking disputes the applicants’ arguments.
83 It should be recalled that, with regard to budgetary matters, the obligation of transparency, which is the corollary of the principle of equal treatment, is essentially intended to preclude any risk of favouritism or arbitrariness on the part of the budgetary authority. The obligation of transparency implies that all the conditions and detailed rules of the grant award procedure must be drawn up in a clear, precise and unequivocal manner, inter alia, in the call for proposals. Accordingly, all the information relevant for the purpose of a sound understanding of the call for proposals must be made available as soon as possible to all the operators who may be interested in a procedure for awarding grants in order, first, to enable all reasonably well-informed and normally diligent applicants to understand their precise scope and to interpret them in the same manner and, second, to enable the budgetary authority actually to verify whether the proposed projects meet the selection and award criteria previously announced. Therefore, any undermining of equality of opportunity and of the principle of transparency constitutes an irregularity invalidating the award procedure (see, to that effect, judgment of 15 April 2011, IPK International v Commission , T‑297/05, EU:T:2011:185, paragraphs 124 and 125 and the case-law cited).
84 Furthermore, Article 29(3) of the Horizon Europe Regulation provides that the evaluation process must be designed to avoid conflicts of interest and bias and that the transparency of the evaluation criteria and of the proposal scoring method must be guaranteed.
85 In the present case, the call for proposals sets out the requirements for centres of excellence, then refers to the relevant documents setting out the applicable general and specific conditions. It is stated in Section 5 of the call for proposals that award criteria, scoring and thresholds are described in Part D of the general annexes to the work programme. As regards submission and evaluation processes for proposals, the applicable rules are described in Part F of the General Annexes and in the online manual entitled ‘Funding and Tenders’ for EU Funding Programmes 2021-2027 (‘the online manual’). Section 7 states that the specific conditions for the call for proposals are described in the 2022 work plan. There is then a list of the relevant documents for the call for proposals, including the application form, and additional documents including the general annexes to the work programme, the Horizon Europe Programme Guide and the online manual.
86 In the first place, with regard to the evaluation of the ‘excellence’ criterion, Part D of the general annexes to the work programme states that the evaluation will take account, inter alia, of the quality of open science practices, including sharing and management of research outputs and engagement of citizens, civil society and end users where appropriate.
87 It was thus required, for the ‘excellence’ criterion, that the proposal address the ‘quality of open science practices’.
88 In the evaluation report, the EuroHPC Joint Undertaking considered that open science practices were addressed, but not adequately described in relation to data management. Furthermore, the EuroHPC Joint Undertaking took the view that the statements were generic, that options for possible data ontologies and metadata were not adequately specified and that the proposal did not give sufficient attention to possible decision paths.
89 The relevant documents mentioned in the call for proposals include the Horizon Europe Programme Guide, adopted by the Commission on 17 June 2021, and the application form.
90 According to its introduction, the Horizon Europe Programme Guide contains detailed guidance on the structure, budget and political priorities of the programme and also includes information on how to prepare proposals. It follows that that guide is a purely explanatory document and the interpretation it gives of the proposal form and its annexes, and of the conditions for the calls for proposals, is not binding on interested parties.
91 The application form contains a template for producing the technical description of the proposal. That form gives applicants guidance on how to complete each section of the prescribed template and to address the required aspects.
92 Although the explanations given in the Horizon Europe Programme Guide and in the application form regarding the criteria taken into account in the evaluation of proposals do not have binding force, those documents nevertheless form part of the context in which the call for proposals was launched, since they are intended to provide, in particular, explanations and suggestions regarding important aspects of proposals.
93 According to the Horizon Europe Programme Guide, applicants are expected to integrate into their proposal both mandatory open science practices and recommended practices. The guide states that applicants will have to provide concrete information on how they plan to comply with the mandatory open science practices. Failure to sufficiently address this will result in a lower evaluation score. A clear explanation of how they will adopt recommended practices, as appropriate for their projects, will result in a higher evaluation score. It is also stated that applicants generating or reusing data must present, in no more than one page, their data management plans.
94 In addition, the application form states that applicants must, first, describe how appropriate open science practices are implemented as an integral part of the proposed methodology and, second, show how the choice of practices and their implementation are adapted to the nature of the work, in a way that will increase the chances of the project delivering on its objectives. The application form also clarifies the notion ‘open science’ and gives examples of open science practices, including open access to research outputs, in particular data.
95 It follows that, since the quality of open science practices is evaluated for the ‘excellence’ and ‘quality and efficiency of the implementation’ criteria and the different documents in the call for proposals, in particular the Horizon Europe Programme Guide, state that research data management practices must be addressed in the proposal, the EuroHPC Joint Undertaking did examine the manner in which the open science aspect was addressed in relation to data management in accordance with the ‘excellence’ criterion and the clarifications included in the guide.
96 In the second place, with regard to the evaluation of the ‘impact’ criterion, the applicants consider, first, that the call for proposals did not require identification of the impact of the project on the targeted medical areas.
97 It must be held, however, that the call for proposals provides that proposals must clearly identify the targeted applications and related codes, including their user basis and the global impact in their domain.
98 In that regard, it is clear from the proposal that the target groups for the applications to be developed in the course of the project are, in particular, medical and biomedical experts and practitioners.
99 To that extent, the EuroHPC Joint Undertaking did not evaluate the proposal in the light of a new criterion but, on the contrary, in the light of one of the requirements laid down in the call for proposals, taking account of the areas targeted by that proposal.
100 Second, the applicants submit that the call for proposals did not require the connection between selected applications and community or market needs to be indicated in the proposal.
101 It should be noted that it is stated in the call for proposals that ‘targeted applications should be relevant for communities of HPC users as well as for future EuroHPC JU systems to be acquired’.
102 Thus, in order to evaluate the impact of the project and, more precisely, the relevance of applications for communities of users, it was necessary to determine the needs of that community and, accordingly, the way in which the development of those applications in the project could meet those needs.
103 Consequently, the connection between selected applications and community or market needs formed part of the evaluation of the impact and the relevance of targeted applications for communities of high-performance computing users. The EuroHPC Joint Undertaking did not therefore apply a criterion which was not envisaged in the call for proposals, contrary to the claim made by the applicants.
104 Lastly, the applicants claim that the call for proposals did not require a consortium agreement to be concluded at the proposal submission stage or intellectual property rights issues to be regulated in detail in their proposal.
105 It should be noted, first, that it is not evident from the evaluation report that the EuroHPC Joint Undertaking evaluated the abovementioned aspects negatively by reason of the absence of a consortium agreement on intellectual property management. The EuroHPC Joint Undertaking thus did not expect the consortium to define the arrangements for intellectual property management in a consortium agreement at the proposal stage. That argument therefore has no factual basis.
106 Second, it is clear from the Horizon Europe Programme Guide that, at the stage of forming the consortium, before submitting the proposal, applicants must already pay attention to eventual and expected results, ownership issues and the associated intellectual property rights with a view to disseminating and exploiting the results efficiently. Furthermore, the Horizon Europe Programme Guide states that in the proposal applicants must outline their strategy for the management of intellectual property, including intended protection measures, if relevant, and how these would be used to support exploitation.
107 Furthermore, it is stated in the Horizon Europe Programme Guide that the consortium agreement sets the framework for successful project implementation and results exploitation, including intellectual property management, and is meant to settle as far as possible all issues that might hamper the smooth and seamless cooperation of the different actors for the different parts of the project. It is also stated that the consortium agreement should in principle be negotiated and concluded before signing the grant agreement.
108 It is thus envisaged that the framework for intellectual property management will be formally set by the consortium agreement, which, as the applicants claim, does not have to be concluded at the proposal submission stage, but in principle before the grant agreement is signed.
109 Nevertheless, having regard to paragraph 106 above, although those issues should have been defined in the consortium agreement, applicants in the call for proposals did not have to wait for that agreement to be concluded to consider intellectual property issues and to outline their strategy in the proposal.
110 The applicants cannot therefore claim that the EuroHPC Joint Undertaking relied on requirements which were not provided for in the call for proposals, taking the view, in the evaluation report, that intellectual property, exploitation rights and confidentiality issues were not adequately addressed in the proposal and finding that management of intellectual property was delegated to the consortium agreement.
111 In the third place, with regard to the ‘quality and efficiency of the implementation’ criterion, the applicants claim that the call for proposals does not requirement the involvement of health professionals in the development of high-performance computing applications.
112 It should be noted that Part D of the general annexes to the work programme provides that, for the ‘quality and efficiency of the implementation’ criterion, the evaluation will take account of the capacity and role of each participant and the extent to which the consortium as a whole brings together the necessary expertise.
113 In addition, the call for proposals requires that proposals clearly demonstrate that all partners in the consortium have a significant and justified role, including appropriate deliverables under their responsibility which cover their specific contributions.
114 The EuroHPC Joint Undertaking considered in the evaluation report that the consortium brought together different highly qualified interdisciplinary partners with complementary expertise as required for the proposed activities, but that there was insufficient involvement, for a centre of excellence with a medical focus, of hospital and medical scientists.
115 Contrary to the claim made by the applicants, the EuroHPC Joint Undertaking did not criticise the consortium for failing to envisage the direct involvement of health professionals in the development of high-performance computing applications, but made a general finding that there was insufficient involvement of hospital and medical scientists in the project.
116 Bringing together the necessary expertise on the part of all the members of the consortium is one of the elements taken into account in the evaluation of the ‘quality and efficiency of the implementation’ criterion, as was stated in paragraph 112 above.
117 The EuroHPC Joint Undertaking did not therefore introduce a new criterion, but simply held, in essence, that the consortium did not bring together sufficient expertise in the medical sector.
118 It follows from the foregoing that, contrary to the claim made by the applicants, the EuroHPC Joint Undertaking did assess the merits of the proposal by reference only to the criteria set out in Part D of the general annexes to the work programme and the requirements laid down in the call for proposals.
119 Consequently, the present plea in law must be rejected.
(c) The first plea in law, alleging a manifest error of assessment
(1) Preliminary observations
120 It should be noted, first of all, that proposals submitted under the Horizon Europe programme in response to the call for proposals were evaluated by the EuroHPC Joint Undertaking in relation to the criteria set out in Part D of the general annexes to the work programme and in the call for proposals. Proposals were evaluated in the light of the award criteria ‘excellence’, ‘impact’ and ‘quality and efficiency of the implementation’.
121 Under the heading ‘Score and weighting’ in Part D of the general annexes to the work programme, it is stated that each criterion is scored out of 5 and that the threshold for individual criteria is 3 points. The overall threshold, applying to the sum of the three individual scores, is 10 points. Proposals that pass the individual threshold and the overall threshold are considered for funding, within the limits of the available call budget. Other proposals are rejected.
122 Consequently, each of the grounds given to justify the award of a score below 3 points would be, on its own, sufficient as a basis for the decision not to accept the proposal with a view to funding under the Horizon Europe programme (see, to that effect, judgments of 14 February 2008, Provincia di Imperia v Commission , T‑351/05, EU:T:2008:40, paragraph 37, and of 16 December 2020, Balti Gaas v Commission and INEA , T‑236/17 and T‑596/17, not published, EU:T:2020:612, paragraph 147). In those circumstances, that decision should be annulled, in principle, only if each of those grounds is unlawful. In that event, an error or other illegality which affects only one of those pillars of reasoning cannot be sufficient to justify the annulment of the contested decision since that error or illegality could not have had a decisive influence as regards the refusal of funding under that programme (see, to that effect, judgments of 10 June 2020, Spliethoff’s Bevrachtingskantoor v Commission , T‑564/15 RENV, not published, EU:T:2020:252, paragraph 62, and of 16 December 2020, Balti Gaas v Commission and INEA , T‑236/17 and T‑596/17, not published, EU:T:2020:612, paragraph 147).
123 Second, with regard to the application of the award criteria, it follows from settled case-law that, as regards the comparative evaluation of various projects in complex scientific and technical fields, judicial review is limited to verifying whether there has been a breach of procedural rules and of the obligation to state reasons, material inaccuracy of the facts, a manifest error of assessment or misuse of powers (see judgment of 18 September 2024, Safran Aircraft Engines v Commission , T‑617/22, not published, EU:T:2024:629, paragraph 110 and the case-law cited).
124 The assessment of the scientific excellence of a project proposal by a panel of reviewers belongs to the category of complex assessments within the meaning of the case-law referred to in paragraph 123 above (see, to that effect, judgments of 25 November 2020, BMC v Clean Sky 2 Joint Undertaking , T‑71/19, not published, EU:T:2020:567, paragraph 74 and the case-law cited, and of 29 March 2023, Universität Bremen v REA , T‑660/19 RENV, not published, EU:T:2023:170, paragraph 35).
125 As regards the evaluation by the EuroHPC Joint Undertaking of a project proposal like that in the present case, which involves complex technical evaluations requiring the use of external experts, it should be noted that considerations similar to those set out in paragraph 124 above can be taken into account as regards the discretion enjoyed by it. The Court’s review of the evaluation of the proposal is therefore limited to verifying whether there is a manifest error of assessment.
126 According to the case-law, in order to establish that an institution has made a manifest error in the assessment of complex facts such as to justify the annulment of an act, the evidence adduced by the applicant must be sufficient to make the factual assessments used in that act implausible. Subject to that review of plausibility, it is not the Court’s role to substitute its assessment of complex facts for that made by the author of that decision. Consequently, a plea alleging the existence of a manifest error of assessment must be rejected if, despite the evidence adduced by the applicant, the assessment in question may still be accepted as true or valid (see judgments of 25 November 2020, BMC v Clean Sky 2 Joint Undertaking , T‑71/19, not published, EU:T:2020:567, paragraph 76 and the case-law cited, and of 29 March 2023, Universität Bremen v REA , T‑660/19 RENV, not published, EU:T:2023:170, paragraph 36 and the case-law cited).
127 Thus, the only situation in which the assessments, which form a decision taken on the basis of complex facts, are capable of being examined by the Court is that in which the applicant alleges that the factual assessments at issue are implausible (judgments of 25 November 2020, BMC v Clean Sky 2 Joint Undertaking , T‑71/19, not published, EU:T:2020:567, paragraph 77, and of 29 March 2023, Universität Bremen v REA , T‑660/19 RENV, not published, EU:T:2023:170, paragraph 37).
128 Against that background, the fact that an applicant whose proposal has not been selected in a call for proposals under the Horizon Europe programme disagrees with the assessments of the experts who examined its proposal, except where it claims that those assessments are implausible, constitutes a challenge which goes beyond the limited review by the Court (see, to that effect, judgments of 25 November 2020, BMC v Clean Sky 2 Joint Undertaking , T‑71/19, not published, EU:T:2020:567, paragraph 78, and of 29 March 2023, Universität Bremen v REA , T‑660/19 RENV, not published, EU:T:2023:170, paragraph 38).
129 It is in the light of the above considerations that the present plea in law must be examined.
(2) The merits of the plea in law
130 The applicants claim, in essence, that the EuroHPC Joint Undertaking made manifest errors of assessment in the evaluation of the three award criteria. By making those errors, the EuroHPC Joint Undertaking infringed the principle of good administration laid down in Article 41 of the Charter of Fundamental Rights of the European Union.
131 The EuroHPC Joint Undertaking disputes the arguments put forward by the applicants.
132 With regard to the ‘excellence’ criterion, first, the applicants claim, in essence, that the EuroHPC Joint Undertaking made an incorrect reading of the proposal by committing factual errors and ignoring elements of the proposal. They thus complain that the EuroHPC Joint Undertaking held, on the one hand, that the proposal had a major weakness in that it did not convincingly demonstrate that the main objective of the project was exascale or that the work of the project required exascale computing power and, on the other hand, that the work described tended to focus on high-throughput computing. The applicants submit that the proposal convincingly demonstrates the need for exascale computing.
133 The applicants assert that the primary objective of the proposal is to promote innovation in the field of modelling and computational biomedicine through the optimisation of lighthouse applications for upcoming exascale architectures. The applicants explain that in the proposal they described three generic patterns for implementing lighthouse applications: monolithic, coupled, and ensemble and workflow (‘Pattern 3’). According to the proposal, the consortium envisaged developing seven applications: HemeLB, ExaDG (monolithic), Alya, HemoCell, IMPECCABLE (coupled), BAC and PhysiBoSS (Pattern 3).
134 They submit that, contrary to the statements made in the evaluation report, the implementation of Pattern 3 is a way to achieve exascale. Furthermore, the performance measures of workflows in Pattern 3 are more involved than merely reporting the number of jobs per unit time, unlike high-throughput computing. In their view, it is therefore incorrect to reduce Pattern 3 and, accordingly, the proposal to high-throughput computing. The applicants note, moreover, that the proposal does not mention ‘high-throughput computing’.
135 In that regard, the applicants state that the EuroHPC Joint Undertaking launched a new call for proposals in which workflows are mentioned as being one of the ways to achieve exascale. Furthermore, the world’s first exascale supercomputer, Frontier, uses inter alia Pattern 3.
136 Moreover, in the context of the fourth plea in law, alleging infringement of the principle of equality, the applicants mentioned various proposals selected for funding where approaches similar to that described in the consortium’s proposal were found by the experts to be convincing and capable of achieving exascale.
137 In addition, by stating that significant parts of the project focused on high-throughput computing, the EuroHPC Joint Undertaking ignored the other two patterns presented in the proposal, namely the monolithic and coupled patterns, covering several applications which are not high-throughput computing applications.
138 Lastly, the applicants maintain that there are inconsistencies in the evaluation report in so far as, on the one hand, the experts concluded that significant parts of the project focused on high-throughput computing and, on the other, they stated that that limited the impact to the field of computational fluid dynamics. In their view, those two conclusions are contradictory and incorrect, since computational fluid dynamics is recognised as a monolithic application, which should be distinguished from high-throughput computing.
139 It should be recalled that Section 5 of the call for proposals states that the award criteria are described in Part D of the general annexes to the work programme. The aspects to be taken into consideration by the experts for the ‘excellence’ criterion include the clarity and pertinence of the project’s objectives, the extent to which the proposal goes beyond the state of the art and demonstrates innovative potential, the relevance of the proposed methodology, including the underlying concepts, models, assumptions, interdisciplinary approaches, consideration of the gender dimension in research and innovation content, and the quality of open science practices, including sharing and management of research outputs and engagement of citizens, civil society and end users where appropriate.
140 In the evaluation report, the experts from the EuroHPC Joint Undertaking considered that the proposal did not convincingly demonstrate that its main objective was exascale. They stated that significant parts of the project focused on high-throughput computing and that the need for exascale in relation to this was not convincingly justified. In their view, that was therefore a major weakness of the proposal.
141 In that regard, it is apparent from the description in the call for proposals that proposals had to identify clearly the exascale lighthouse applications addressed and demonstrate convincingly their exascale capabilities and needs. Proposals also had to articulate the scientific challenges addressed by the proposals and the reasons why exascale performance was needed.
142 The applicants claim, in essence, that the EuroHPC Joint Undertaking equated Pattern 3 with high-throughput computing. However, as is rightly pointed out by the EuroHPC Joint Undertaking, it is not evident from the evaluation report that that was the basis for its assessment. Indeed, the EuroHPC Joint Undertaking does not dispute that that pattern may lead to exascale. It considers, however, that significant parts of the work focus on high-throughput computing and that the need for exascale computing was not convincingly justified. It must therefore be stated that the argument relied on is based on the incorrect premiss that the EuroHPC Joint Undertaking equated the proposed Pattern 3 with high-throughput computing. Consequently, that argument must be rejected as having no factual basis.
143 Thus, given that it is not apparent from the evaluation report that the EuroHPC Joint Undertaking equated one or more patterns with high-throughput computing, it must be examined whether, regardless of the pattern of the applications, the consortium demonstrated that the main objective of the proposal was to achieve exascale performance.
144 In that regard, it is for the applicants to adduce sufficient evidence to render implausible the assessment made by the EuroHPC Joint Undertaking that significant parts of the project focused on high-throughput computing and that the proposal did not justify the need to utilise exascale computing power for the development of applications.
145 The applicants assert that Pattern 3 permits exascale to be achieved and should be not associated with high-throughput computing. It must be stated, however, that the applicants simply state in general, with imprecise references to certain pages of the proposal, that many other applications will be implemented in Pattern 3 and that that mode will support the most applications at the exascale level, without indicating the applications to which they are referring. Thus, the applicants did not clarify which applications mentioned in the proposal permit exascale to be achieved.
146 Second, the applicants cite proposals to which grants were awarded in the call for proposals, such as the ExaWorks project, for which the workflow model was accepted as a valid method for reaching exascale. The applicants highlight the similarities between the workflows implemented in that project and the IMPECCABLE and BAC Pattern 3 applications under the proposal.
147 Although the applicants do identify some similarities in the software and workflows implemented by the ExaWorks project and those covered by their proposal, those aspects are not, however, sufficient to render implausible the EuroHPC Joint Undertaking’s assessment that the proposal does not demonstrate convincingly the exascale needs of the project.
148 The applicants also claim, in essence, that the evaluation report ignores the other two patterns mentioned in paragraph 137 above. Certain applications, such as the HemeLB and Alya applications, which are covered by the monolithic and coupled patterns respectively, are not high-throughput computing applications. Those applications, which are already operational, serve as benchmark applications on exascale or near-exascale systems.
149 It should be observed, however, that the applicants do not demonstrate that the applications mentioned in paragraph 148 above permit exascale to be reached and require high-performance computing power. Furthermore, although they submit that computational fluid dynamics is a monolithic application which must be distinguished from high-throughput computing, they do not indicate the specific applications to which they are referring and, above all, how those applications permit exascale to be reached.
150 Lastly, the applicants assert that the consortium is running several of the applications referred to in the proposal on the first exascale supercomputer called Frontier in the INCITE programme funded by the United States Department of Energy, referring to Annexes C.6 and C.7, the admissibility of which is challenged by the EuroHPC Joint Undertaking. They argue that this confirms that the codes for the applications referred to in the proposal also run on exascale machines. However, it is sufficient to note that the applicants do not specify which applications, of those which would be developed in the course of the project, are run on the Frontier supercomputer. Furthermore, and without it being necessary to rule on the admissibility of Annexes C.6 and C.7, it should be recalled that it is not for the General Court to supplement the arguments put forward by the applicants in their written submissions, by searching for and identifying, in the annexes, evidence capable of supporting their claims (see, to that effect, judgment of 13 February 2025, Commission and Others v Carpatair , C‑244/23 P to C‑246/23 P, EU:C:2025:87, paragraph 77).
151 It follows from the foregoing that the applicants have not submitted sufficient evidence to render implausible the assessment that the need for exascale performance had not been demonstrated convincingly in the proposal. Accordingly, the EuroHPC Joint Undertaking cannot be considered to have made a manifest error of assessment in applying that requirement laid down in the call for proposals.
152 Second, the applicants claim, in essence, that the EuroHPC Joint Undertaking made a manifest error of assessment in considering, in the first place, that the proposal did not specifically present the expected improvements of the applications which are intended to use exascale computing power, in terms of performance, scalability and energy efficiency, in so far as it provided only generic key performance indicators (‘KPIs’) and, in the second place, that the proposal did not adequately address metrics to measure objectives in terms of improvement of personalised healthcare.
153 In the first place, the applicants claim that only some KPIs in the proposal were taken into account by the EuroHPC Joint Undertaking. They list 13 indicators which, in their view, are not generic and, moreover, relate to all the elements aspects identified in the evaluation report, namely the performance (KPI 1, 2, 4, 5), scalability (KPI 3, 8, 9) and energy efficiency (KPI 10) of the applications in question. Furthermore, they submit that the proposal identified in detail the expected outcomes and impact of the ‘escalation’ of each application.
154 In this case, it is clear from the evaluation report that the EuroHPC Joint Undertaking considered that the proposal did not specifically present the expected improvements towards exascale in the applications which are intended to use exascale computing power, in terms of performance, scalability and energy efficiency, in so far as it provided only generic KPIs.
155 In that regard, the call for proposals requires centres of excellence to include clear KPIs on the optimal use of current and emergent high-performance computing technologies, allowing the assessment of progress towards the objectives, in terms of both outputs and ultimate impact.
156 It must be held that, although the applicants challenge the EuroHPC Joint Undertaking’s assessment, they simply state that certain KPIs are related to performance and others to scalability and energy efficiency, without explaining how those KPIs made it possible to measure the expected improvements of the applications towards exascale. Furthermore, the applicants merely affirm that the 13 indicators mentioned in the application are not generic, without substantiating their claim. That argument does not therefore render implausible the assessment in the evaluation report that only generic KPIs were provided in the proposal.
157 In the second place, the applicants claim that the proposal identifies in detail the expected outcomes and impact of the escalation of the application. They also submit that the proposal indicates the research results, the outcome, the dissemination and exploitation measures, the target groups and the expected concrete impacts.
158 It is sufficient to note in that regard that the applicants merely refer to Section 2.1.2 of the proposal, without explaining how the proposal clearly determined the expected improvements of the applications towards exascale, in terms of performance, scalability and energy efficiency.
159 Consequently, the arguments put forward by the applicants do not render implausible the EuroHPC Joint Undertaking’s assessment mentioned in paragraph 154 above.
160 Third, the applicants claim, in essence, that, in its evaluation of the ‘excellence’ criterion, the EuroHPC Joint Undertaking, on the one hand, made a manifest error of assessment and, on the other hand, contradicted the evaluation of the ‘impact’ criterion by considering in the evaluation report that procedures for obtaining access to data sets were not sufficiently addressed in the methodology.
161 They maintain that the proposal identifies difficulties in access to patient data as a potential obstacle in the project, but states that that obstacle would be overcome by using, for most of the work, public data and anonymised synthetic datasets. The applicants add that, given the technical ambitions of the project, expecting the earliest exascale simulations solely to utilise only individual patient data is unrealistic. They specify that patient data constitutes only part of the data intended for use within the project.
162 The applicants also submit that the proposal contained a detailed and adequate treatment of required technical improvements and data access, presenting, in particular, data management based on the ‘FAIR’ principles (Findability, Accessibility, Interoperability, Reusability).
163 Furthermore, with regard to the alleged contradiction between the assessment of the ‘excellence’ criterion and the assessment of the ‘impact’ criterion, the applicants assert that the EuroHPC Joint Undertaking considered, in the evaluation of that criterion, that the mitigating measures identified in the proposal to overcome potential barriers, including the issue of access to data, are reasonable and their management is reasonably addressed.
164 It should be noted that the call for proposals requires centres of excellence to present demonstrable advances of the targeted high-performance computing applications towards highly scalable, optimised flagship codes and exascale performance (both computing and extreme data). That includes developing, maintaining, porting, optimising (if needed redesigning) and scaling of high-performance computing application codes, addressing the full scientific or industrial workflows, particularly data aspects, testing and validating codes and quality assurance.
165 On the one hand, the applicants simply make an imprecise reference to pages 18 to 20 of the proposal, in which the applications to be developed in the course of the project are presented, without specifying the detail contained in the proposal regarding the required technical improvements to achieve exascale performance level for those applications and workflows. It should be stated in that respect, as was recalled in paragraph 150 above, that it is not for the General Court to supplement the arguments put forward by the applicants in their written submissions, by searching for and identifying, in the annexes, evidence on which the action could be based.
166 Furthermore, the applicants merely refer to certain sections of the proposal relating to access to data, without identifying precisely the information contained in the proposal. They simply state in particular, referring to Section 2.1.3 of the proposal, entitled ‘Requirements and potential barriers’, that the consortium had identified the obstacle relating to access to sufficient patient data, but that it would be remedied by using public data which is readily available for download or can be obtained in a short time. Although that difficulty in access to patient data had actually been identified in the proposal, the applicants did not put forward sufficient arguments to render implausible the fact that the proposed methodology did not properly take into account the limitations in terms of access to data.
167 On the other hand, as regards the alleged contradiction between the evaluation of the ‘excellence’ criterion and the assessment of the ‘impact’ criterion, the evaluation report states, in respect of the latter criterion, that the proposal identifies a number of potential barriers and framework conditions that could have an influence on the extent to which the expected impacts of the project may be achieved, which are reasonable and whose management is reasonably addressed.
168 It follows that, contrary to the assertion made by the applicants, the EuroHPC Joint Undertaking did not take the view in the evaluation report that the difficulties in access to data were reasonable or that their management was reasonably addressed. It must therefore be stated that the applicants distort the content of the contested decision to which the evaluation report is annexed.
169 The applicants’ arguments are not therefore such as to call into question the plausibility of the assessment contained in the evaluation report with regard to consideration of difficulties in access to data in the methodology of the proposal.
170 Fourth, the applicants claim, in essence, that the EuroHPC Joint Undertaking made a manifest error of assessment by affirming that the co-design approaches did not take into account that some applications presented have very different requirements.
171 According to the applicants, the proposal makes clear that co-design relates to all applications, without being confined to any specific application. The approaches proposed are thus generic and applicable to all kinds of applications, but also sufficiently flexible to be adapted to the specific requirements of each application. The applicants state in that regard that co-design is in an initial development stage and that there are not yet established methods or standard protocols which can be deployed specifically for a given application.
172 The proposal also mentions the consortium’s experience in co-design.
173 The applicants claim that, of the five proposed work packages, Work Package 4 provides considerable detail on the different applications which they propose to execute on exascale platforms. Furthermore, that work package is the second most heavily staffed work package, which shows the importance attached to co-design in the project.
174 In addition, the proposal covers all the main aspects to be optimised during co-design in relation to all the applications included in the proposal. Lastly, the consortium brings together all the necessary expertise in the software and hardware domains.
175 It should be stated that in the evaluation report the EuroHPC Joint Undertaking took the view that co-design approaches were proposed, but that they were not fully convincing. The EuroHPC Joint Undertaking noted, by way of example, that the applications presented had very different requirements. It also considered that it was not sufficiently explained to which application the co-design approaches related and under which technical requirements the work with hardware developers would take place.
176 According to the description in the call for proposals, proposals should be inherently committed to co-design activities to ensure that future high-performance computing architectures are well suited for the applications and their users. The call for proposals further requires from centres of excellence involvement in co-design activities (hardware, software, codes), including collaboration with high-performance computing vendors and the identification of applications relevant to the development of European high-performance computing technologies towards exascale and collaboration with European initiatives.
177 It must be held that the applicants simply state that the proposed co-design approaches are generic and can thus be adapted to each application, without substantiating that claim. They do not demonstrate that the EuroHPC Joint Undertaking made a manifest error of assessment in taking the view that the proposal had not specified to which application the co-design approaches related or explained sufficiently under which technical requirements the work with hardware developers would take place.
178 The applicants refer to page 32 of the proposal, which presents the pathways towards impact envisaged by the proposal. The applicants do not, however, cite any specific passages of the proposal in support of their argument.
179 Lastly, according to the case-law cited in paragraphs 126 to 128 above, it is not the Court’s role to substitute its assessment of complex facts for that made by the author of the decision in question. Thus, the only situation in which the assessments of the evaluation report are capable of being examined by the Court is that in which the applicants allege that the factual assessments at issue are implausible. However, the applicants’ argument that, first, co-design is in an initial development stage and, second, there are not yet specific methods for any given application goes beyond the limited review by the Court. Furthermore, the applicants have not substantiated their argument on that point.
180 The applicants have not therefore adduced any evidence to render implausible the assessment mentioned in paragraph 175 above.
181 Fifth, the applicants challenge the assessment made in the evaluation report of the description of the baseline of the research. The applicants claim that Section 1.1.2 of the proposal precisely discusses the baseline of the research. In the first place, the proposal notes that, in the existing literature, there are instances of generic human-scale 1D vascular models. However, the system proposed by the consortium would aim at achieving realistic simulations of entirely digital twins of the human body. In the second place, the baseline is clearly defined and discussed in pages 4 to 8 of the proposal, in which the current performance data for all applications to be developed under the proposed project are provided.
182 It should be observed that in the evaluation report the EuroHPC Joint Undertaking took the view that the proposal would not advance significantly beyond the state of the art, that the baseline was not adequately discussed and that the proposal did not convincingly demonstrate a significant advance from the point of view of evolution towards exascale. The evaluation report also states that the proposal does not convincingly describe what scientific breakthroughs can be achieved through that work.
183 Part D of the general annexes to the work programme expressly provides that the aspects to be taken into account in the evaluation of the ‘excellence’ criterion include the ambition of the project and the extent to which it goes beyond the state of the art.
184 Although it is clear from Section 1.1.2. of the proposal that the consortium described the state of the research which could at that stage develop human-scale 1D vascular models and stated that the ultimate aim of the CompBioMed3 project was to achieve realistic simulations of digital twins of the human body, it should be noted that the applicants do not specify in that section how the consortium will be able to achieve the envisaged scientific breakthroughs, allowing them to go beyond the state of the art.
185 Thus, the applicants’ arguments do not render implausible the assessment in the evaluation report mentioned in paragraph 182 above.
186 Sixth, the applicants claim, in essence, that the assessment in the evaluation report that open science practices are not adequately described in relation to data management is incorrect. According to the applicants, Sections 1.2.6 and 1.2.7 of the proposal describe in detail the open science practices and related policies which the consortium proposes to develop. Furthermore, Work Package 5 explains the proposed data management approach in detail.
187 Part D of the general annexes to the work programme provides that the evaluation of the ‘excellence’ criterion will take into account inter alia the quality of open science practices (see also paragraphs 87 to 94 above).
188 It should be noted that the applicants simply state that the proposal describes in detail the proposed open science practices and related policies, which is not disputed in the evaluation report, and the data management approach, referring to certain passages of that proposal, without identifying whether there is a connection between those practices and data management. It follows that the applicants have not adduced any evidence to render implausible the assessment that open science practices were not adequately described in relation to data management.
189 Accordingly, the arguments alleging a manifest error of assessment regarding the ‘excellence’ criterion must be rejected as being unfounded. Consequently, the EuroHPC Joint Undertaking cannot be considered to have committed any infringement of the principle of good administration in that context. For the reasons set out in paragraph 122 above, according to which each of the grounds given to justify the award of a score below 3 points is, on its own, sufficient as a basis for the decision not to accept the proposal with a view to funding under the Horizon Europe programme, the present plea in law must therefore be rejected in its entirety, without it being necessary to examine the arguments concerning a manifest error of assessment regarding the other award criteria.
(d) The fourth plea in law, alleging infringement of the principle of equal treatment
190 The applicants claim that the EuroHPC Joint Undertaking committed an infringement of the principle of equal treatment and non-discrimination by failing to treat equally the different applicants which submitted a proposal for funding, when they were in the same situation. They submit that, in the case of other proposals selected by the EuroHPC Joint Undertaking, the experts considered Pattern 3 to be a valid method for reaching exascale, whereas in the contested decision the experts wrongly took the view that the proposal focused on high-throughput computing and failed to consider Pattern 3.
191 The applicants claim that that difference in treatment had consequences in the assessment of the proposal against the three award criteria.
192 As regards the ‘excellence’ and ‘quality and efficiency of the implementation’ criteria, the applicants assert that if the EuroHPC Joint Undertaking had considered Pattern 3 as a viable option to reach exascale, as is evident from the evaluation reports for other proposals, it would certainly have concluded that the proposal did have a significant advance towards exascale and contained the appropriate KPIs.
193 As far as the ‘impact’ criterion is concerned, the applicants submit that the evaluation report concluded that the proposal was limited to computational fluid dynamics, while a substantial component of the work was based on ensembles and workflows applied in numerous other areas of computational biomedicine. In their view, the failure to consider the variety of computing patterns proposed to reach exascale adversely affected the evaluation of the proposal.
194 The EuroHPC Joint Undertaking disputes the applicants’ arguments.
195 As a preliminary point, it should be noted that the principle of non-discrimination relied on by the applicant must be understood as referring to the principle of equal treatment (see judgment of 12 March 2020, XB v ECB , T‑484/18, not published, EU:T:2020:90, paragraph 49 and the case-law cited).
196 It follows from settled case-law that the principle of equal treatment requires that comparable situations are not treated differently and that different situations are not treated in the same way unless such treatment is objectively justified (see, to that effect, judgment of 11 October 2023, DCM Film Distribution v Commission , T‑760/21, not published, EU:T:2023:612, paragraph 100 and the case-law cited).
197 A breach of the principle of equal treatment as a result of different treatment thus presupposes that the situations concerned are comparable, having regard to all the elements which characterise them. It must be recalled that the elements which characterise different situations, and hence their comparability, must in particular be determined and assessed in the light of the subject matter and purpose of the European Union act which makes the distinction in question. The principles and objectives of the field to which the act relates must also be taken into account (see judgment of 11 October 2023, DCM Film Distribution v Commission , T‑760/21, not published, EU:T:2023:612, paragraph 101 and the case-law cited).
198 The principle of equal treatment between candidates in a procedure for the award of grants also implies an obligation of transparency in order to enable verification that it has been complied with (see judgment of 11 October 2023, DCM Film Distribution v Commission , T‑760/21, not published, EU:T:2023:612, paragraph 102 and the case-law cited).
199 It that regard, it must be observed, first, that the plea in law alleging infringement of the principle of equal treatment is based on a false premiss. As was stated in paragraph 142 above, in the contested decision the EuroHPC Joint Undertaking did not take the view that Pattern 3 was not a valid method for reaching exascale. In this case, the applicants do not specify how the extracts from the evaluation reports for the proposals cited in the application demonstrate that the EuroHPC Joint Undertaking took the view that Pattern 3 was a valid method for reaching exascale and that it had thus evaluated the consortium’s proposal differently in that regard from the other successful proposals.
200 Second, it must be stated that the applicants have not demonstrated that their proposal was comparable to the successful proposals mentioned by them, simply affirming that all the applicants were in the same situation. The applicants merely assert that in the evaluation report for the successful proposals contained in Annexes A.16 to A.18 the EuroHPC Joint Undertaking had taken the view that Pattern 3 was a valid method for reaching exascale, contrary to the contested decision. They do not, however, describe the successful proposals nor claim that those proposals are equivalent to theirs, merely stating that all the applicants are in the same situation.
201 It should be noted that, in the context of the plea alleging an infringement of the principle of equal treatment as a general principle of EU law, the burden of proving that the situations in which the applicant finds itself are comparable with its competitors rests on the applicant (see judgment of 10 June 2020, Spliethoff’s Bevrachtingskantoor v Commission , T‑564/15 RENV, not published, EU:T:2020:252, paragraph 97 and the case-law cited).
202 The applicants do not therefore demonstrate that the EuroHPC Joint Undertaking committed an infringement of the principle of equal treatment and the principle of non-discrimination. Consequently, the present plea in law and thus the claim for annulment directed against the contested decision must be rejected.
2. The claim for annulment directed against the decision rejecting the request for review
203 With regard to the decision rejecting the request for review, the applicants rely on two pleas in law, alleging, first, infringement of the obligation to state reasons and, second, infringement of Article 30(2) of the Horizon Europe Regulation. The second plea in law should be examined first.
(a) The second plea in law, alleging infringement of Article 30 (2) of the Horizon Europe Regulation
204 The applicants complain that the EuroHPC Joint Undertaking took the view that the errors claimed in the request for an evaluation review fell outside the scope of the review procedure. They assert that the term ‘procedural aspects’ in Article 30(2) of the Horizon Europe Regulation cannot be interpreted too strictly. In that respect, they maintain inter alia that, according to the online manual, procedural aspects also encompass factual errors, manifest errors of assessment and abuse of powers, namely lack of coherence between scores and comments, and lack or inadequate reasoning of the conclusions. The applicants thus submit that the arguments put forward in their request for review do not merely relate to the merits of the proposal.
205 The EuroHPC Joint Undertaking disputes the applicants’ arguments.
206 Article 30(1) of the Horizon Europe Regulation provides that an applicant may request an evaluation review if it considers that the applicable evaluation procedure has not been correctly applied to its proposal. It is provided in Article 30(2) of that regulation that only the procedural aspects of an evaluation may be the subject of a request for an evaluation review and that the evaluation of the merits of a proposal are not to be the subject of an evaluation review.
207 It is apparent from the wording of Article 30 of the Horizon Europe Regulation that the evaluation review procedure for the grant proposal may relate only to the procedural regularity of that evaluation, and not to the merits of the proposal.
208 It should be noted that in Section 5 of the call for proposals it is stated that the submission and evaluation processes are described in Part F, entitled ‘Procedure’, of the general annexes to the work programme and in the online manual. It is clear in that regard from the introduction in that manual that it is limited to procedural aspects of grant procedures and explaining the functionalities of the IT tools behind the ‘Funding and Tenders’ electronic portal.
209 It must therefore be held that the ‘procedural aspects’ which may be the subject of a request for review on the basis of Article 30(1) of the Horizon Europe Regulation relate to the procedure as described in the documents mentioned in paragraph 208 above.
210 The applicants’ argument concerning the online manual, mentioned in paragraph 204 above, is not such as to call that finding into question. As the applicants assert, the online manual states that applicants may raise factual errors, manifest errors of assessment or abuse of powers.
211 However, the online manual does not constitute a legally binding text which is capable of derogating from the binding rule laid down in Article 30(2) of the Horizon Europe Regulation. As was mentioned in paragraph 208 above, the online manual is intended to explain the procedural aspects of grants to applicants for a grant and its beneficiaries.
212 Accordingly, the fact that applicants are informed that they can raise procedural irregularities, factual errors, manifest errors of assessment or abuse of powers cannot result in an extension of the scope of the review procedure defined by Article 30 of the Horizon Europe Regulation.
213 It should be noted that, in the request for an evaluation review, the consortium submitted, first, that the contested decision was based on incorrect facts, that it ignored a large part of the proposal and that the EuroHPC Joint Undertaking made manifest errors of assessment. Second, the consortium complains that the EuroHPC Joint Undertaking committed an infringement of the obligation to state reasons. In particular, the consortium alleged, in essence, that the EuroHPC Joint Undertaking also failed to provide details in the statement of reasons regarding certain assessments contained in the evaluation report. Third, the consortium took the view that the EuroHPC Joint Undertaking had assessed the proposal against sub-criteria which had not been envisaged in the call for proposals, in breach of the obligation of transparency. To that end, the consortium presented a table in which it disputed the assessments made by the EuroHPC Joint Undertaking’s experts concerning the three award criteria.
214 Following the response to its application for access to documents, the consortium sent the EuroHPC Joint Undertaking, on 22 December 2022, a supplement to its request for an evaluation review, by which it claimed infringement of the principle of equal treatment and non-discrimination.
215 It must be stated that the arguments put forward in the request for an evaluation review did not relate to procedural aspects of the evaluation procedure, but only to the merits of the evaluation. As regards, in particular, the claim of infringement of the principle of transparency, the arguments put forward in the request for an evaluation review actually relate to the merits of the proposal and the application thereto of the criteria for the call for proposals, rather than the procedure in itself.
216 The present plea in law must therefore be rejected.
(b) The first plea in law, alleging infringement of the obligation to state reasons
217 The applicants claim that the EuroHPC Joint Undertaking infringed the obligation to state reasons on the ground that, in the decision rejecting the request for review, it did not give them sufficient information to allow the consortium to know whether the decision at issue may have been vitiated by an error enabling its validity to be challenged. First, as regards the infringement of the obligation to state reasons claimed in that request, the EuroHPC Joint Undertaking fails to explain why it believes that the statement of reasons is sufficient. Second, as regards the alleged infringement of the principle of transparency, the applicants submit that the EuroHPC Joint Undertaking fails to demonstrate why it believes that the award criteria were duly applied. Third, the applicants claim that the EuroHPC Joint Undertaking fails to address specifically why it believed that there was no different treatment of the proposal. In particular, the EuroHPC Joint Undertaking’s claim that it follows a consistent approach in the agreement on the scores and in the comments throughout the different evaluation phases is merely a general answer which does not allow the reasons why there was equal treatment to be understood.
218 The EuroHPC Joint Undertaking disputes the applicants’ arguments.
219 It should be noted that, by the decision rejecting the request for review, the EuroHPC Joint Undertaking took the view that the evaluation procedure had been carried out in accordance with the applicable rules. Consequently, the EuroHPC Joint Undertaking held that the evaluation result stood and that there would be no further evaluation. The EuroHPC Joint Undertaking took the view that the alleged factual errors claimed concerned conclusions in the evaluations on the merits of the proposal and therefore fell outside the scope of the review procedure under Article 30(2) of the Horizon Europe Regulation. Furthermore, the EuroHPC Joint Undertaking considered that sufficient reasoning had been provided and that the evaluators had not made manifest errors of assessment or factual errors. In addition, the EuroHPC Joint Undertaking took the view that the experts had merely applied the award criteria in the call for proposals in accordance with the principle of transparency. Lastly, the EuroHPC Joint Undertaking held that equal treatment was ensured by following the rules and procedures under the Horizon Europe Regulation throughout the different evaluation phases.
220 Accordingly, the decision rejecting the request for review discloses, in a clear and unequivocal fashion, the reasons why the request for review was rejected.
221 Moreover, in so far as the review procedure laid down in Article 30(2) of the Horizon Europe Regulation relates only to procedural aspects of the evaluation, and not to the relevance of the proposal, as was pointed out in paragraphs 206 to 216 above, while the arguments raised by the consortium in its request for review relate only to the merits of the assessments made by the experts, and not to procedural aspects of the evaluation, the EuroHPC Joint Undertaking was able, without infringing the obligation to state reasons in the review procedure, to refrain from responding to the consortium’s complaints relating to the merits of the assessments made by the experts (see, to that effect, judgment of 25 November 2020, BMC v Clean Sky 2 Joint Undertaking , T‑71/19, not published, EU:T:2020:567, paragraph 55).
222 It must be held that the statement of reasons in the decision rejecting the request for review meets the requisite legal standard having regard to the requirements stemming from the case-law cited in paragraphs 61 to 64 above, in particular in the light of the legal rules governing the evaluation review procedure. Consequently, the present plea in law and, therefore, the claim for annulment directed against the decision rejecting the request for review must be rejected.
223 It follows from all the foregoing that the action must be dismissed in its entirety.
IV. Costs
224 Under Article 134(1) of the Rules of Procedure, the unsuccessful party is to be ordered to pay the costs if they have been applied for in the successful party’s pleadings.
225 Since the applicants have been unsuccessful, they must be ordered jointly and severally to bear their own costs and to pay the costs incurred by the EuroHPC Joint Undertaking, in accordance with the form of order sought by it.
On those grounds,
THE GENERAL COURT (Ninth Chamber)
hereby:
1. Dismisses the action;
2. Orders Radical-Consulting UG, Surf BV and Bayerische Akademie der Wissenschaften jointly and severally to pay the costs.
Truchot | Sampol Pucurull | Perišin
Delivered in open court in Luxembourg on 4 February 2026.
V. Di Bucci | M. van der Woude
Registrar | President
* Language of the case: English.