ARE-bench
Agents continue an existing research project from a real user request. A judge checks the deliverable against the request and earlier workspace requirements.
The authors grade 75 tasks with 1,173 binary checklist items. A task passes only when every applicable item passes.
- Released
- November 2026
- Kind
- AI research tasks
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Agent task
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Terminal
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- A host
- In views
- Agents
- Strongest model tested
- Claude Opus 5.5 #5 on the capability index
- Last model added
- September 2026
- Scores refreshed
- November 21st, 2026
GPT-6.1 Sol leads with 66.2% among models released by September 2026
Best score per model from the benchmark's paper, read on November 21st, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6.1 Terra
- and 14 more
- 1 GPT-6.1 Sol 66.2%
- 2 Claude Opus 5.5 56.9%
- 3
DeepSeek V4.1 Flash 39.6%
- 4
Qwen 3.8 Max 37.8%
- 5 GLM-5.3-Flash 30.7%
- 6 Kimi K3 28%
- 7
Qwen 3.8 27B 25.8%
Source: The ARE-bench paper
| # | Model | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6.1 Sol | 66.2% ±5.4 | Sep 2026 | Official leaderboard | |
| 2 | Claude Opus 5.5 | 56.9% ±4.3 | Sep 2026 | Official leaderboard | |
| 3 | | 39.6% ±3.8 | Sep 2026 | Official leaderboard | |
| 4 | | 37.8% ±4.3 | Aug 2026 | Official leaderboard | |
| 5 | GLM-5.3-Flash | 30.7% ±7.1 | Aug 2026 | Official leaderboard | |
| 6 | Kimi K3 | 28% ±2.3 | Jul 2026 | Official leaderboard | |
| 7 | | 25.8% ±0.8 | Aug 2026 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The ARE-bench paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official ARE-bench leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
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-6.2 Astra 96.1%
- 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. Kimi K3 0.71
- KernelBench In use Replace PyTorch operators with faster custom GPU kernels. Scored on correctness and speedup.
MiniMax-M3 28.8%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (7 entries).
- Pass Models scored 7 models we track.
- Pass Newest score The newest is from September 2026, for GPT-6.1 Sol.
- Note Frontier models Only 2 of the top 10 have a score. The strongest tested is Claude Opus 5.5 (#5).
- Pass Headroom The best is 66.2%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 7 rows for models we track at the last read.
- Pass Last read November 21st, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.