Benchmark Almanac

McStasBench

Claude Sonnet 5 43.8% leads since September 2026
Not in model cards

Agents design neutron scattering instruments. A simulator checks whether their designs reproduce published measurements. The authors score two additional instruments separately.

The maintainers publish scores for seven models across 16 tasks with and without tools. They update the tables from recorded trials.

Released
September 2026
Kind
Lab and data work
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Actions, Code
Interaction
Agentic
Environment
McStas
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
In views
Agents, Robotics
Strongest model tested
Claude Sonnet 5 #35 on the capability index
Last model added
July 2026
Scores refreshed
November 20th, 2026
7
Models scored on its own leaderboard

Claude Sonnet 5 leads with 43.8% 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. Every run is in the next chart.

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:

  1. 1 Claude Sonnet 5 oneshot 43.8%
  2. 2 Gemini 3.6 Flash oneshot 37.5%
  3. 3 Claude Haiku 4.5 main 25%
  4. 4 Qwen3-32B main 18.8%
  5. 5 Llama 4 Maverick oneshot 12.5%
  6. 6 Gemini 3.5 Flash-Lite main 12.5%
  7. 7 Qwen3-8B main 6.25%

Source: Official McStasBench leaderboard

#ModelScoreSettingDateSource
1 Claude Sonnet 5 43.8% oneshot Jun 2026 Official leaderboard
2 Gemini 3.6 Flash 37.5% oneshot Jul 2026 Official leaderboard
3 Claude Haiku 4.5 25% main Oct 2025 Official leaderboard
4 Qwen3-32B 18.8% main Apr 2025 Official leaderboard
5 Llama 4 Maverick 12.5% oneshot Apr 2025 Official leaderboard
6 Gemini 3.5 Flash-Lite 12.5% main Jul 2026 Official leaderboard
7 Qwen3-8B 6.25% main Apr 2025 Official leaderboard

Scores at each effort level

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Claude Sonnet 5
main oneshot
43.8%
Gemini 3.6 Flash
main oneshot
37.5%
Claude Haiku 4.5
main
25%
Qwen3-32B
oneshot main
18.8%
Llama 4 Maverick
main oneshot
12.5%
Gemini 3.5 Flash-Lite
oneshot main
12.5%
Qwen3-8B
oneshot main
6.25%

Source: Official McStasBench leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Sonnet 5 oneshot 43.8%
Gemini 3.6 Flash oneshot 37.5%
Claude Sonnet 5 main 31.3%
Claude Haiku 4.5 main 25%
Claude Haiku 4.5 oneshot 25%
Gemini 3.6 Flash main 25%
Qwen3-32B main 18.8%
Llama 4 Maverick oneshot 12.5%
Gemini 3.5 Flash-Lite main 12.5%
Qwen3-8B main 6.25%
Gemini 3.5 Flash-Lite oneshot 6.25%
Llama 4 Maverick main 0%
Qwen3-32B oneshot 0%
Qwen3-8B oneshot 0%

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

Source: Official McStasBench leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Claude Sonnet 5 Claude Haiku 4.5 Qwen3-32B Llama 4 Maverick 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Claude Sonnet 5 Claude Haiku 4.5

Source: Official McStasBench 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

All 203 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 All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (7 entries).
  • Pass Models scored 7 models we track.
  • Pass Newest score The newest is from July 2026, for Gemini 3.5 Flash-Lite.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Claude Sonnet 5 (#35).
  • Pass Headroom The best is 43.8%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 14 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.

How we check