Benchmark Almanac

AutoResearchExam

Claude Fable 5.1 0.60 leads since Sep 2026 Not in model cards

Agents get 24 hours on a CPU or GPU machine to improve solutions to open-ended ML research tasks through experiments. The score is the area under the improvement curve on a hidden test set, rewarding both quality and speed.

29 tasks over seven areas, including model training, data curation, AI safety and interpretability. Released 2026-09-09. The hidden-test metric is meant to catch agents that overfit the visible data.

Made by Bespoke Labs Released Sep 2026 Machine learning engineering All tasks public Last model added Sep 2026 Scores refreshed 2 Nov 2026
Level
Research frontier Researchers, on open problems or tasks that take days
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
bespokelabsai/AutoResearchExam 35 stars Leaderboard benchmarks.bespokelabs.ai
9
Models scored on its own leaderboard

Claude Fable 5.1 leads with 0.60 among models released by Sep 2026

Best score per model from the benchmark's own leaderboard, read on 2 Nov 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in Sep 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 0.60 Claude Fable 5.1
  2. 0.60 GPT-6 Astra
  3. 0.58 Claude Opus 5
  4. 0.51 GPT-5.6 Sol
  5. 0.49 Grok 4.6
  6. 0.48 Gemini 3.8 Flash
  7. 0.43 Kimi K3
  8. 0.42 Qwen 3.8 Max
  9. 0.38 Muse Spark 1.3
  1. 1 Claude Fable 5.1 0.60
  2. 2 GPT-6 Astra 0.60
  3. 3 Claude Opus 5 0.58
  4. 4 GPT-5.6 Sol 0.51
  5. 5 Grok 4.6 0.49
  6. 6 Gemini 3.8 Flash 0.48
  7. 7 Kimi K3 0.43
  8. 8 Qwen 3.8 Max 0.42
  9. 9 Muse Spark 1.3 0.38
#ModelScoreSettingDateSource
1 Claude Fable 5.1 0.60 Sep 2026 Official leaderboard
2 GPT-6 Astra 0.60 Sep 2026 Official leaderboard
3 Claude Opus 5 0.58 Jul 2026 Official leaderboard
4 GPT-5.6 Sol 0.51 Jul 2026 Official leaderboard
5 Grok 4.6 0.49 Aug 2026 Official leaderboard
6 Gemini 3.8 Flash 0.48 Sep 2026 Official leaderboard
7 Kimi K3 0.43 Jul 2026 Official leaderboard
8 Qwen 3.8 Max 0.42 Aug 2026 Official leaderboard
9 Muse Spark 1.3 0.38 Sep 2026 Official leaderboard

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic
Own leaderboard Not tested since Sep 2026
Claude Fable 5.1 0.60

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard
0.40 0.50 0.60 0.40 0.50 0.60

No lab model card reports it yet

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.

More in AI research

All 83 in AI research
  • MLS-Bench In use Agent must invent one algorithmic change in an ML research scaffold whose gain transfers across settings. Scored relative to human baselines. Claude Fable 5.1 50.3%
  • PaperBench In use Agent replicates ICML 2024 papers from scratch. LLM judge grades against author-written hierarchical rubrics. Qwen 3.8 Max 93%
  • MLE-bench In use Kaggle-style ML engineering competitions. Agent trains models and is scored against competition leaderboards. GPT-6 Astra 93.8%
  • PostTrainBench 1.2 In use CLI agent gets a base model, one H100 and a time budget to post-train it for a target benchmark. Scored on normalized gain. Claude Fable 5.1 44.6%
  • KernelBench In use Replace PyTorch operators with faster custom GPU kernels. Scored on correctness and speedup. MiniMax-M3 28.8%
  • OpenAI PRs Only in system cards Agents replicate internal OpenAI pull requests from a pre-PR branch and a prompt. Graded by hidden unit tests as pass@1. GPT-5.2 55%