CoBench (Anthropic)
500 real internal Anthropic R&D problems. Percentage solved in one attempt tracks AI R&D acceleration.
v2.1 scores not comparable with earlier CoBench versions.
- Made by
- Anthropic
- Kind
- Labs' internal R&D tests
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Agent task
- Tasks
- Kept private
- Strongest model tested
- Claude Opus 5.5 #4 on the capability index
- Last in a system card
- September 2026
- Last model added
- September 2026
Claude Opus 5.5 leads with 55.8%
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
- 1 Claude Opus 5.5 55.8%
Source: Labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | 55.8% | Sep 2026 | Lab model card |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Labs’ model cards
Labs report it only in system cards
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.
One system card reports it
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Sep 22, 2026 | Claude Opus 5.5 system card | 55.8 | CoBench 2.1, share of 500 internal R&D problems solved in one attempt, API safety filter off. Not comparable with earlier CoBench versions (2.3.4.1, Figure 2.3.4.1.A) | AI R&D risk |
More in AI research
- PaperBench In use Agent replicates ICML 2024 papers from scratch. LLM judge grades against author-written hierarchical rubrics.
Qwen 3.8 Max 93% - 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%
- 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%
- MLE-bench In use Kaggle-style ML engineering competitions. Agent trains models and is scored against competition leaderboards. GPT-6 Astra 93.8%
- KernelBench In use Replace PyTorch operators with faster custom GPU kernels. Scored on correctness and speedup.
MiniMax-M3 28.8% - PMPP-Hard In use Agents write GPU kernels for 69 tasks that must be correct, and on 48 of them also run within 1.25 times the reference time. The score is the area under the solved-tasks curve over a 10-hour budget, out of 690 task-hours.
Inkling 2 Small 494.3