# Automated AI Research on MLE-Bench official (full) Source: [Automated AI Research on MLE-Bench official (full)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:08:06.306Z Evaluation of model performance on the full version of the MLE-Bench dataset, focusing on metrics related to submission validity and achievement tiers (Bronze, Silver, Gold). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Automated AI Research | | Dataset (including any stated split) | MLE-Bench official (full) | | Metric | Valid Submission Rate | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 13 of 13 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Valid Submission Rate](): Higher is better; unit: PERCENT. - [Above Median Rate](): Higher is better; unit: PERCENT. - [Bronze Tier Achievement](): Higher is better; unit: PERCENT. - [Silver Tier Achievement](): Higher is better; unit: PERCENT. - [Gold Tier Achievement](): Higher is better; unit: PERCENT. - [Any Medal Achievement Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Valid Submission Rate | Result date | Report paper | | --- | --- | --- | --- | | MARS | 98.7 | Not recorded | 6982d27da8b01e76cc8e921d | | MARS+ | 98.7 | Not recorded | 6982d27da8b01e76cc8e921d | | AIRA-dojo | 98.2 | Not recorded | 6982d27da8b01e76cc8e921d | | Famou-Agent | 96.9 | Not recorded | 6982d27da8b01e76cc8e921d | | InternAgent | 96.4 | Not recorded | 6982d27da8b01e76cc8e921d | | ML-Master 2.0 | 95.6 | Not recorded | 6982d27da8b01e76cc8e921d | | MARS | 94.2 | Not recorded | 6982d27da8b01e76cc8e921d | | ML-Master | 93.3 | Not recorded | 6982d27da8b01e76cc8e921d | | AIDE | 84.4 | Not recorded | 6982d27da8b01e76cc8e921d | | AIRA-dojo | 83.6 | Not recorded | 6982d27da8b01e76cc8e921d | | AIDE | 82.7 | Not recorded | 6982d27da8b01e76cc8e921d | | R&D-Agent | 53.3 | Not recorded | 6982d27da8b01e76cc8e921d | | Leeroo | 50.7 | Not recorded | 6982d27da8b01e76cc8e921d | ### MARS Valid Submission Rate: 98.7. Result ID: 699609f35698665ec441dd62. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Gemini-3-Pro-Preview","Evaluation Environment":"Controlled Evaluation in Our Environment"}. This record does not assert independent reproduction. ### MARS+ Valid Submission Rate: 98.7. Result ID: 699609f35698665ec441dd63. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Gemini-3-Pro-Preview","Evaluation Environment":"Controlled Evaluation in Our Environment"}. This record does not assert independent reproduction. ### AIRA-dojo Valid Submission Rate: 98.2. Result ID: 699609f35698665ec441dd60. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Gemini-3-Pro-Preview","Evaluation Environment":"Controlled Evaluation in Our Environment"}. This record does not assert independent reproduction. ### Famou-Agent Valid Submission Rate: 96.9. Result ID: 699609f35698665ec441dd5a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Gemini-2.5-Pro","Evaluation Environment":"Official MLE-Bench Leaderboard"}. This record does not assert independent reproduction. ### InternAgent Valid Submission Rate: 96.4. Result ID: 699609f35698665ec441dd59. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Deepseek-R1","Evaluation Environment":"Official MLE-Bench Leaderboard"}. This record does not assert independent reproduction. ### ML-Master 2.0 Valid Submission Rate: 95.6. Result ID: 699609f35698665ec441dd5c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Deepseek-V3.2-Speciale","Evaluation Environment":"Official MLE-Bench Leaderboard"}. This record does not assert independent reproduction. ### MARS Valid Submission Rate: 94.2. Result ID: 699609f35698665ec441dd61. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Gemini-2.5-Pro","Evaluation Environment":"Controlled Evaluation in Our Environment"}. This record does not assert independent reproduction. ### ML-Master Valid Submission Rate: 93.3. Result ID: 699609f35698665ec441dd57. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Deepseek-R1","Evaluation Environment":"Official MLE-Bench Leaderboard"}. This record does not assert independent reproduction. ### AIDE Valid Submission Rate: 84.4. Result ID: 699609f35698665ec441dd5d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Gemini-2.5-Pro","Evaluation Environment":"Controlled Evaluation in Our Environment"}. This record does not assert independent reproduction. ### AIRA-dojo Valid Submission Rate: 83.6. Result ID: 699609f35698665ec441dd5f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Gemini-2.5-Pro","Evaluation Environment":"Controlled Evaluation in Our Environment"}. This record does not assert independent reproduction. ### AIDE Valid Submission Rate: 82.7. Result ID: 699609f35698665ec441dd5e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Gemini-3-Pro-Preview","Evaluation Environment":"Controlled Evaluation in Our Environment"}. This record does not assert independent reproduction. ### R&D-Agent Valid Submission Rate: 53.3. Result ID: 699609f35698665ec441dd58. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-5","Evaluation Environment":"Official MLE-Bench Leaderboard"}. This record does not assert independent reproduction. ### Leeroo Valid Submission Rate: 50.7. Result ID: 699609f35698665ec441dd5b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Gemini-3-Pro-Preview","Evaluation Environment":"Official MLE-Bench Leaderboard"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.