MooseBench
Models write input files for MOOSE, a multiphysics simulator. The grader reconstructs each file’s governing equations and compares them with the intended physics.
The authors publish scores for six models on the same 220 cases under two repair methods. The full inputs and scoring map require a request.
- Made by
- Zhenghan Song, Yulong Liu, and colleagues
- Released
- May 2026
- Kind
- Scientific code
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Code
- Tasks
- Kept private
- Measures
- Capability
- Language
- English
- Interaction
- Multi-turn
- Refresh
- One fixed set
- Run by
- A host
- Strongest model tested
- GPT-5.4 #28 on the capability index
- Last model added
- April 2026
- Scores refreshed
- November 20th, 2026
Claude Sonnet 4.6 leads with 0.85 among models released by April 2026 Claude Sonnet 4.6 leads with 0.85 among models released by April 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran on its own with several settings, we keep its best. Every run is in the next chart. Agents other companies built around a model get a pale shade of that model's lab color. Gray marks blends and entries not in our model list, such as fine-tunes and older models. None of these count toward leaders. Agents built around a model, such as Gemini 3.1 Flash-Lite + DeepSeek-V4-Flash, are their own entries. We don't count them as the model's score. This view shows only models we track, each at its best.
The newest model with a score here came out in April 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 16 more
- 1 Claude Sonnet 4.6 PDE-Reg 0.85
- 2 GPT-5.4 PDE-Reg 0.84
- 3
Gemini 3.1 Flash-Lite PDE-Reg 0.83
- 4
DeepSeek-V4-Flash PDE-Reg 0.82
- 5 Claude Haiku 4.5 PDE-Reg 0.81
- 6 Gemini 3.1 Flash-Lite + DeepSeek-V4-Flash PDE-Refine, extractor: DeepSeek V4 Flash 0.79
- 7 DeepSeek-V4-Flash + Gemini 3.1 Flash-Lite PDE-Refine, extractor: Gemini 3.1 Flash Lite 0.78
- 8 GPT-5.4 + DeepSeek-V4-Flash PDE-Refine, extractor: DeepSeek V4 Flash 0.78
- 9 DeepSeek-V4-Flash + GPT-5.4 PDE-Refine, extractor: GPT-5.4 0.77
- 10 GPT-4.1 mini PDE-Reg 0.70
Source: Official MooseBench leaderboard
- 1 Claude Sonnet 4.6 PDE-Reg 0.85
- 2 GPT-5.4 PDE-Reg 0.84
- 3
Gemini 3.1 Flash-Lite PDE-Reg 0.83
- 4
DeepSeek-V4-Flash PDE-Reg 0.82
- 5 Claude Haiku 4.5 PDE-Reg 0.81
- 6 GPT-4.1 mini PDE-Reg 0.70
Source: Official MooseBench leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 0.85 | PDE-Reg | Feb 2026 | Official leaderboard |
| 2 | GPT-5.4 | 0.84 | PDE-Reg | Mar 2026 | Official leaderboard |
| 3 | | 0.83 | PDE-Reg | Mar 2026 | Official leaderboard |
| 4 | | 0.82 | PDE-Reg | Apr 2026 | Official leaderboard |
| 5 | Claude Haiku 4.5 | 0.81 | PDE-Reg | Oct 2025 | Official leaderboard |
| 6 | Gemini 3.1 Flash-Lite + DeepSeek-V4-Flash | 0.79 | PDE-Refine, extractor: DeepSeek V4 Flash | Mar 2026 | Official leaderboard |
| 7 | DeepSeek-V4-Flash + Gemini 3.1 Flash-Lite | 0.78 | PDE-Refine, extractor: Gemini 3.1 Flash Lite | Apr 2026 | Official leaderboard |
| 8 | GPT-5.4 + DeepSeek-V4-Flash | 0.78 | PDE-Refine, extractor: DeepSeek V4 Flash | Mar 2026 | Official leaderboard |
| 9 | DeepSeek-V4-Flash + GPT-5.4 | 0.77 | PDE-Refine, extractor: GPT-5.4 | Apr 2026 | Official leaderboard |
| 10 | GPT-4.1 mini | 0.70 | PDE-Reg | Apr 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 0.85 | PDE-Reg | Feb 2026 | Official leaderboard |
| 2 | GPT-5.4 | 0.84 | PDE-Reg | Mar 2026 | Official leaderboard |
| 3 | | 0.83 | PDE-Reg | Mar 2026 | Official leaderboard |
| 4 | | 0.82 | PDE-Reg | Apr 2026 | Official leaderboard |
| 5 | Claude Haiku 4.5 | 0.81 | PDE-Reg | Oct 2025 | Official leaderboard |
| 6 | GPT-4.1 mini | 0.70 | PDE-Reg | 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.
Source: Official MooseBench leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Claude Sonnet 4.6 | PDE-Reg | 0.85 | |||
| GPT-5.4 | PDE-Reg | 0.84 | |||
| PDE-Reg | 0.83 | ||||
| PDE-Reg | 0.82 | ||||
| Claude Sonnet 4.6 | PDE-Refine | 0.82 | |||
| Docs+PDE-Reg | 0.82 | ||||
| Claude Haiku 4.5 | PDE-Reg | 0.81 | |||
| GPT-5.4 | PDE-Refine | 0.80 | |||
| Claude Sonnet 4.6 | SpecGen | 0.80 | |||
| PDE-Refine, extractor: Gemini 3.1 Flash Lite | 0.80 | ||||
| GPT-5.4 | PDE-Refine, extractor: GPT-5.4 | 0.79 | |||
| Gemini 3.1 Flash-Lite + DeepSeek-V4-Flash | PDE-Refine, extractor: DeepSeek V4 Flash | 0.79 | |||
| DeepSeek-V4-Flash + Gemini 3.1 Flash-Lite | PDE-Refine, extractor: Gemini 3.1 Flash Lite | 0.78 | |||
| GPT-5.4 + DeepSeek-V4-Flash | PDE-Refine, extractor: DeepSeek V4 Flash | 0.78 | |||
| PDE-Refine | 0.78 | ||||
| DeepSeek-V4-Flash + GPT-5.4 | PDE-Refine, extractor: GPT-5.4 | 0.77 | |||
| GPT-5.4 | SpecGen | 0.77 | |||
| PDE-Refine, extractor: DeepSeek V4 Flash | 0.76 | ||||
| GPT-5.4 | Exec-Repair+Reg | 0.76 | |||
| GPT-5.4 | Direct | 0.76 | |||
| Claude Sonnet 4.6 | Exec-Repair+Reg | 0.75 | |||
| Claude Sonnet 4.6 | Direct | 0.74 | |||
| Claude Haiku 4.5 | Exec-Repair+Reg | 0.73 | |||
| GPT-4.1 mini | PDE-Reg | 0.70 | |||
| DocsDirect | 0.67 | ||||
| Exec-Repair+Reg | 0.67 | ||||
| SpecGen | 0.67 | ||||
| Docs+ExecRepair | 0.66 | ||||
| Exec-Repair+Reg | 0.63 | ||||
| Direct | 0.60 | ||||
| GPT-4.1 mini | Exec-Repair+Reg | 0.35 |
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.
Source: Official MooseBench leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Claude Sonnet 4.6 | PDE-Reg | 0.85 | |||
| GPT-5.4 | PDE-Reg | 0.84 | |||
| PDE-Reg | 0.83 | ||||
| PDE-Reg | 0.82 | ||||
| Claude Sonnet 4.6 | PDE-Refine | 0.82 | |||
| Docs+PDE-Reg | 0.82 | ||||
| Claude Haiku 4.5 | PDE-Reg | 0.81 | |||
| GPT-5.4 | PDE-Refine | 0.80 | |||
| Claude Sonnet 4.6 | SpecGen | 0.80 | |||
| PDE-Refine, extractor: Gemini 3.1 Flash Lite | 0.80 | ||||
| GPT-5.4 | PDE-Refine, extractor: GPT-5.4 | 0.79 | |||
| PDE-Refine | 0.78 | ||||
| GPT-5.4 | SpecGen | 0.77 | |||
| PDE-Refine, extractor: DeepSeek V4 Flash | 0.76 | ||||
| GPT-5.4 | Exec-Repair+Reg | 0.76 | |||
| GPT-5.4 | Direct | 0.76 | |||
| Claude Sonnet 4.6 | Exec-Repair+Reg | 0.75 | |||
| Claude Sonnet 4.6 | Direct | 0.74 | |||
| Claude Haiku 4.5 | Exec-Repair+Reg | 0.73 | |||
| GPT-4.1 mini | PDE-Reg | 0.70 | |||
| DocsDirect | 0.67 | ||||
| Exec-Repair+Reg | 0.67 | ||||
| SpecGen | 0.67 | ||||
| Docs+ExecRepair | 0.66 | ||||
| Exec-Repair+Reg | 0.63 | ||||
| Direct | 0.60 | ||||
| GPT-4.1 mini | Exec-Repair+Reg | 0.35 |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official MooseBench 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 MooseBench 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 Coding
- SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
- LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests.
DeepSeek-V4-Pro 93.5%
- SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
- DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
- SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases.
Seed2.1 Pro 59.8%
- Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance.
DeepSeek V4.1 Flash 3,471
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for DeepSeek-V4-Flash. Every benchmark
Catalog record 2 pass
- Pass Task access The tasks are private.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (10 entries).
- Pass Models scored 6 models we track, plus 4 other entries on its leaderboard.
- Warning Newest score The newest came 7 months ago, in April 2026, for DeepSeek-V4-Flash.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.4 (#28).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 31 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.