Frontier Agentic STEM
Agents produce scientific results in a terminal. Task verifiers recompute the expected results from the underlying data.
Turing reports outcomes for five models on 805 tasks.
- Made by
- Turing
- Released
- August 2026
- Kind
- Research science
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- Kept private
- Measures
- Capability
- Subject
- Other natural sciences
- Interaction
- Agentic
- Environment
- Terminal
- Refresh
- One fixed set
- Run by
- A host
- In views
- Agents
- Strongest model tested
- Claude Opus 5 #11 on the capability index
- Last model added
- July 2026
- Scores refreshed
- November 20th, 2026
GPT-5.6 Sol leads with 55.1% among models released by July 2026
Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in July 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 Astra
- and 14 more
- 1 GPT-5.6 Sol high 55.1%
- 2 Claude Opus 5 high 54.4%
- 3 GPT-5.5 medium 45.5%
- 4 Claude Opus 4.8 high 43.2%
- 5
Gemini 3.5 Flash medium 30.6%
Source: Official Frontier Agentic STEM leaderboard
| # | Model | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.6 Sol | 55.1% ±0.6 | high, Terminus-2 | Jul 2026 | Official leaderboard |
| 2 | Claude Opus 5 | 54.4% ±0.6 | high, Terminus-2 | Jul 2026 | Official leaderboard |
| 3 | GPT-5.5 | 45.5% ±0.6 | medium, Terminus-2 | Apr 2026 | Official leaderboard |
| 4 | Claude Opus 4.8 | 43.2% ±0.6 | high, Terminus-2 | May 2026 | Official leaderboard |
| 5 | | 30.6% ±0.6 | medium, Terminus-2 | May 2026 | Official leaderboard |
Rankings by part
We read 12 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Physics, Biology and Mathematics. GPT-5.6 Sol leads 6 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Physics
5 models we track have a score here.
- 1 GPT-5.6 Sol 39.9%
- 2 Claude Opus 5 39.2%
- 3 GPT-5.5 29.7%
- 4 Claude Opus 4.8 27.9%
- 5
Gemini 3.5 Flash 14.6%
Biology
5 models we track have a score here.
- 1 Claude Opus 5 ↑1 51.3%
- 2 GPT-5.6 Sol ↓1 49.5%
- 3 Claude Opus 4.8 ↑1 45.4%
- 4 GPT-5.5 ↓1 37.7%
- 5
Gemini 3.5 Flash 29.7%
Mathematics
5 models we track have a score here.
- 1 Claude Opus 5 ↑1 72.6%
- 2 GPT-5.6 Sol ↓1 71.4%
- 3 GPT-5.5 63.5%
- 4 Claude Opus 4.8 60.7%
- 5
Gemini 3.5 Flash 48%
Chemistry
5 models we track have a score here.
- 1 Claude Opus 5 ↑1 55%
- 2 GPT-5.6 Sol ↓1 50.9%
- 3 GPT-5.5 47.4%
- 4 Claude Opus 4.8 42.1%
- 5
Gemini 3.5 Flash 33.9%
Mathematics & Theoretical Sciences
5 models we track have a score here.
- 1 GPT-5.6 Sol 74.1%
- 2 Claude Opus 5 62%
- 3 GPT-5.5 59.3%
- 4 Claude Opus 4.8 58.3%
- 5
Gemini 3.5 Flash 36.1%
Computational Chemistry
5 models we track have a score here.
- 1 GPT-5.6 Sol 56.9%
- 2 Claude Opus 5 55.9%
- 3 GPT-5.5 43.1%
- 4 Claude Opus 4.8 41.2%
- 5
Gemini 3.5 Flash 22.5%
Computational and Theoretical Chemistry
5 models we track have a score here.
- 1 Claude Opus 5 ↑1 38.5%
- 2 GPT-5.6 Sol ↓1 28.2%
- 3 Claude Opus 4.8 ↑1 21.8%
- 4 GPT-5.5 ↓1 20.5%
- 5
Gemini 3.5 Flash 7.70%
Computational Physics
5 models we track have a score here.
- 1 GPT-5.6 Sol 62.7%
- 2 Claude Opus 5 58.7%
- 3 GPT-5.5 46.7%
- 4
Gemini 3.5 Flash ↑1 46.7%
- 5 Claude Opus 4.8 ↓1 41.3%
Computational Biology
5 models we track have a score here.
- 1 GPT-5.6 Sol 70.6%
- 2 Claude Opus 4.8 ↑2 64.7%
- 3 Claude Opus 5 ↓1 64.7%
- 4 GPT-5.5 ↓1 62.7%
- 5
Gemini 3.5 Flash 56.9%
Chemistry & Molecular Engineering
5 models we track have a score here.
- 1 GPT-5.5 ↑2 84.6%
- 2 Claude Opus 4.8 ↑2 74.4%
- 3 GPT-5.6 Sol ↓2 66.7%
- 4 Claude Opus 5 ↓2 61.5%
- 5
Gemini 3.5 Flash 33.3%
Data Science
5 models we track have a score here.
- 1 GPT-5.5 ↑2 78.8%
- 2 Claude Opus 5 75.8%
- 3 Claude Opus 4.8 ↑1 69.7%
- 4 GPT-5.6 Sol ↓3 69.7%
- 5
Gemini 3.5 Flash 54.5%
Biophysics
5 models we track have a score here.
- 1 GPT-5.6 Sol 80%
- 2 Claude Opus 5 80%
- 3
Gemini 3.5 Flash ↑2 73.3%
- 4 Claude Opus 4.8 73.3%
- 5 GPT-5.5 ↓2 56.7%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Frontier Agentic STEM leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Frontier Agentic STEM 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 Science and reasoning
- GPQA Diamond In use 198 graduate-level 'Google-proof' multiple-choice questions in biology, chemistry and physics.
Gemini 4 Carbon 97%
- Terminal-Bench-Science In use 70 scientific research-workflow tasks across five science areas, run in Harbor containers. Hidden tests give the resolution rate. GPT-6 Astra 68.1%
- BioMysteryBench In use Bioinformatics problems on anonymized real datasets that require real analysis with tools. Rubric-graded, with human-solvable and human-difficult splits. Claude Fable 5.2 83.7%
- FrontierScience In use Models solve physics, chemistry, and biology problems in an Olympiad track graded by short answer and a Research track graded by 10-point rubrics.
Intern-S2-397B 87%
- HiPhO In use Models solve theory problems from 2024 and 2025 physics olympiad exams, some with diagrams. Official marking schemes grade answers and steps, mapped to medal levels.
Gemini 3 Pro 78.3%
- LABBench2 In use 11 biology research sub-tasks (literature, figures, tables, patents, trials and more) with tool access. Macro-averaged accuracy. GPT-5.2 60.9%
Data audit
All 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access The tasks are private.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (5 entries).
- Pass Models scored 5 models we track.
- Pass Newest score The newest is from July 2026, for Claude Opus 5.
- Note Frontier models None of the top 10 has a score. The strongest tested is Claude Opus 5 (#11).
- Pass Headroom The best is 55.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on labs.turing.com: 5 rows for models we track at the last read.
- Pass Last read November 20th, 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.