Benchmark Almanac

MatTools

Gemini-1.5-Pro (snapshot unspecified) 81.8% tops its leaderboard Qwen2.5-72B 81.6% leads the models we track
Not in model cards

Models answer questions about materials science software and write Python programs to calculate material properties. The authors execute the generated programs.

Made by
Siyu Liu and colleagues
Released
May 2025
Kind
Lab and data work
Level
University
Scored
Checked automatically
Answers
Code
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Code → Text, Code
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Gemini 2.0 Flash Feb 2025 #147 on the capability index
Last model added
February 2025
Scores refreshed
November 20th, 2026
12
Entries in its paper, 4 models we track

Gemini-1.5-Pro (snapshot unspecified) tops the leaderboard with 81.8% Qwen2.5-72B leads with 81.6% among models released by February 2025

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper reports no overall score, so this is the plain mean of code and documentation QA accuracy. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Qwen2.5-72B, at 81.6%. This view shows only models we track, each at its best.

The newest model with a score here came out in February 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Gemini-1.5-Pro (snapshot unspecified) 81.8%
  2. 2 Qwen2.5-72B 81.6%
  3. 3 Qwen2.5-32B-Instruct 81%
  4. 4 Qwen2.5-Coder-32B-Instruct 79.5%
  5. 5 Qwen2.5-14B-Instruct 78.5%
  6. 6 Qwen2.5-Coder-14B-Instruct 78.4%
  7. 7 Gemini 2.0 Flash Feb 2025 78.3%
  8. 8 Gemini-1.5-Flash (snapshot unspecified) 76%
  9. 9 Qwen2.5-7B-Instruct 75.1%
  10. 10 ChemDFM-v1.5-8B 31.3%
  11. 11 ChemLLM-7B-Chat-1_5-DPO 0.15%
  12. 12 Darwin 1.5-7B 0.01%

Source: The MatTools paper

#EntryScoreSettingDateSource
1 Gemini-1.5-Pro (snapshot unspecified) 81.8% Official leaderboard
2 Qwen2.5-72B 81.6% Sep 2024 Official leaderboard
3 Qwen2.5-32B-Instruct 81% Official leaderboard
4 Qwen2.5-Coder-32B-Instruct 79.5% Nov 2024 Official leaderboard
5 Qwen2.5-14B-Instruct 78.5% Official leaderboard
6 Qwen2.5-Coder-14B-Instruct 78.4% Nov 2024 Official leaderboard
7 Gemini 2.0 Flash Feb 2025 78.3% Feb 2025 Official leaderboard
8 Gemini-1.5-Flash (snapshot unspecified) 76% Official leaderboard
9 Qwen2.5-7B-Instruct 75.1% Official leaderboard
10 ChemDFM-v1.5-8B 31.3% Official leaderboard
11 ChemLLM-7B-Chat-1_5-DPO 0.15% Official leaderboard
12 Darwin 1.5-7B 0.01% Official leaderboard
#ModelScoreSettingDateSource
1 Qwen2.5-72B 81.6% Sep 2024 Official leaderboard
2 Qwen2.5-Coder-32B-Instruct 79.5% Nov 2024 Official leaderboard
3 Qwen2.5-Coder-14B-Instruct 78.4% Nov 2024 Official leaderboard
4 Gemini 2.0 Flash Feb 2025 78.3% Feb 2025 Official leaderboard

Rankings by part

We read 2 parts from the benchmark’s paper: Code QA accuracy and Documentation QA accuracy. Each ranks models on one part of the test. Qwen2.5-72B leads Code QA accuracy and Gemini 2.0 Flash leads Documentation QA accuracy. Arrows show how far a model moved from its overall rank among the models we track.

Code QA accuracy

4 models we track have a score here.

  1. 1 Qwen2.5-72B 81.4%
  2. 2 Qwen2.5-Coder-32B-Instruct 79.2%
  3. 3 Qwen2.5-Coder-14B-Instruct 77.4%
  4. 4 Gemini 2.0 Flash Feb 2025 73.8%

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Qwen
Benchmark paper Not tested since May 2025
Qwen2.5-72B 81.6%

Source: The MatTools paper

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
In the benchmark’s paper
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Qwen2.5-72B 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Qwen2.5-72B

Source: Official MatTools 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

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 21 months ago, in February 2025, for Gemini 2.0 Flash. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the benchmark’s paper (12 entries).
  • Pass Models scored 4 models we track, plus 8 other entries on its leaderboard.
  • Warning Newest score The newest came 21 months ago, in February 2025, for Gemini 2.0 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 2.0 Flash (#147).
  • Pass Headroom The best is 81.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 4 rows for models we track and 8 other entries 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