Benchmark Almanac

NeurIPS 2025 Experts Benchmark

Qwen3-Coder-480B-A35B 73.3% leads since December 2025
Not in model cards

Models answer tasks submitted by participants at NeurIPS 2025 in San Diego. Kaggle reports an overall score and individual task outcomes.

Made by
Kaggle
Released
December 2025
Kind
Language reasoning
Level
University
Scored
Settled by events
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Gemini 3 Pro #56 on the capability index
Last model added
December 2025
Scores refreshed
November 20th, 2026
10
Models scored on its own leaderboard

5 models share the top score of 73.3% among models released by December 2025

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 December 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 Qwen3-Coder-480B-A35B 73.3%
  2. 2 Gemini 3 Pro 73.3%
  3. 3 Claude Opus 4.5 73.3%
  4. 4 DeepSeek-V3.2 73.3%
  5. 5 Gemini 3 Flash 73.3%
  6. 6 Gemini 2.0 Flash Feb 2025 66.7%
  7. 7 DeepSeek-V3.1 66.7%
  8. 8 Gemini 2.5 Flash Sep 2025 66.7%
  9. 9 Claude Haiku 4.5 66.7%
  10. 10 Gemma 3 1B 53.3%

Source: Official NeurIPS 2025 Experts Benchmark leaderboard

#ModelScoreSettingReleasedRun onSource
1 Qwen3-Coder-480B-A35B 73.3% Jul 2025 Jan 2026 Official leaderboard
2 Gemini 3 Pro 73.3% Nov 2025 Jan 2026 Official leaderboard
3 Claude Opus 4.5 73.3% Nov 2025 Jan 2026 Official leaderboard
4 DeepSeek-V3.2 73.3% Dec 2025 Jan 2026 Official leaderboard
5 Gemini 3 Flash 73.3% Dec 2025 Jan 2026 Official leaderboard
6 Gemini 2.0 Flash Feb 2025 66.7% Feb 2025 Jan 2026 Official leaderboard
7 DeepSeek-V3.1 66.7% Aug 2025 Jan 2026 Official leaderboard
8 Gemini 2.5 Flash Sep 2025 66.7% Sep 2025 Mar 2026 Official leaderboard
9 Claude Haiku 4.5 66.7% Oct 2025 Jan 2026 Official leaderboard
10 Gemma 3 1B 53.3% Mar 2025 Jan 2026 Official leaderboard

Rankings by part

We read 15 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Generate TikZ 3D Sphere in Cube, List all horse breeds and have an LLM judge and predict nvidia market value. Gemini 2.0 Flash leads 12 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Generate TikZ 3D Sphere in Cube

10 models we track have a score here.

  1. 1 Gemini 2.0 Flash Feb 2025 ↑5 100%
  2. 2 Gemma 3 1B ↑8 100%
  3. 3 Qwen3-Coder-480B-A35B ↓2 100%
  4. 4 DeepSeek-V3.1 ↑3 100%
  5. 5 Gemini 2.5 Flash Sep 2025 ↑3 100%
  6. 6 Claude Haiku 4.5 ↑3 100%
  7. 7 Gemini 3 Pro ↓5 100%
  8. 8 Claude Opus 4.5 ↓5 100%
  9. 9 DeepSeek-V3.2 ↓5 100%
  10. 10 Gemini 3 Flash ↓5 100%

Number one over time

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

Qwen
Own leaderboard Not tested since Dec 2025

Source: Official NeurIPS 2025 Experts Benchmark 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%) Qwen3-Coder-480B-A35B Gemini 2.0 Flash 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Qwen3-Coder-480B-A35B Gemini 2.0 Flash

Source: Official NeurIPS 2025 Experts Benchmark 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 Knowledge

All 332 in Knowledge
  • MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects. Intern-S2-397B 89.8%
  • MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy. Seed1.8 92.3%
  • SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy. Qwen 3.6 Max (Preview) 73.9%
  • WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1. Tencent Hy4 preview 83.9%
  • BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly. Gemma 4 31B IT 74.4%
  • HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 11 months ago, in December 2025, for Gemini 3 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 official leaderboard (10 entries).
  • Pass Models scored 10 models we track.
  • Warning Newest score The newest came 11 months ago, in December 2025, for Gemini 3 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3 Pro (#56).
  • Pass Headroom The best is 73.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on kaggle.com: 10 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