Benchmark Almanac

ReasonZoo

Qwen3-235B-A22B 61% leads since August 2025
Not in model cards

Models solve tasks in logical reasoning, puzzles, and programming. The evaluation compares their answers with task solutions.

Released
August 2025
Kind
Abstract puzzles and logic
Level
University
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text, Code
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
DeepSeek-R1 May 2025 #122 on the capability index
Last model added
May 2025
Scores refreshed
November 20th, 2026
7
Entries in its paper, 4 models we track

Qwen3-235B-A22B leads with 61% among models released by May 2025 Qwen3-235B-A22B leads with 61% among models released by May 2025

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.

The newest model with a score here came out in May 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 Qwen3-235B-A22B MT-TIR 61%
  2. 2 DeepSeek-R1 May 2025, MT-TIR 55.3%
  3. 3 Qwen3-32B PoT 49.9%
  4. 4 Qwen3-8B PoT 43%
  5. 5 ToRL TIT 18.7%
  6. 6 CIR TIT 16.9%
  7. 7 Qwen2.5-Math-7B No tools 4.80%

Source: The ReasonZoo paper

#EntryScoreSettingDateSource
1 Qwen3-235B-A22B 61% MT-TIR Apr 2025 Official leaderboard
2 DeepSeek-R1 May 2025 55.3% MT-TIR May 2025 Official leaderboard
3 Qwen3-32B 49.9% PoT Apr 2025 Official leaderboard
4 Qwen3-8B 43% PoT Apr 2025 Official leaderboard
5 ToRL 18.7% TIT Official leaderboard
6 CIR 16.9% TIT Official leaderboard
7 Qwen2.5-Math-7B 4.80% No tools Official leaderboard
#ModelScoreSettingDateSource
1 Qwen3-235B-A22B 61% MT-TIR Apr 2025 Official leaderboard
2 DeepSeek-R1 May 2025 55.3% MT-TIR May 2025 Official leaderboard
3 Qwen3-32B 49.9% PoT Apr 2025 Official leaderboard
4 Qwen3-8B 43% PoT Apr 2025 Official leaderboard

Rankings by part

We read 9 parts from the benchmark’s paper, each ranking models on one part of the test, such as Number Calculation, Grade School Math and Puzzle. Qwen3-235B-A22B leads 5 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.

Number Calculation

4 models we track have a score here.

  1. 1 DeepSeek-R1 May 2025 ↑1 69.4%
  2. 2 Qwen3-32B ↑1 65.5%
  3. 3 Qwen3-235B-A22B ↓2 58.4%
  4. 4 Qwen3-8B 46%

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
Qwen3-235B-A22B
No tools PoT MT-TIR
61%
DeepSeek-R1
PoT No tools MT-TIR
55.3%
Qwen3-32B
No tools MT-TIR PoT
49.9%
Qwen3-8B
No tools MT-TIR
43%

Source: The ReasonZoo paper

ModelEffortHarnessScoreCost per taskOutput tokens
Qwen3-235B-A22B MT-TIR 61% 14,066
DeepSeek-R1 May 2025 MT-TIR 55.3% 11,284
Qwen3-235B-A22B PoT 52.8% 10,552
Qwen3-32B PoT 49.9% 9,893
Qwen3-32B MT-TIR 47.3% 11,899
Qwen3-235B-A22B No tools 45.8% 9,160
Qwen3-8B PoT 43% 10,855
Qwen3-32B No tools 42.4% 9,160
DeepSeek-R1 May 2025 No tools 41.9% 8,278
Qwen3-8B MT-TIR 41% 15,462
Qwen3-8B No tools 37.6% 11,942
DeepSeek-R1 May 2025 PoT 36.6% 8,310

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
Qwen3-235B-A22B
No tools PoT MT-TIR
61%
DeepSeek-R1
PoT No tools MT-TIR
55.3%
Qwen3-32B
No tools MT-TIR PoT
49.9%
Qwen3-8B
No tools MT-TIR
43%

Source: The ReasonZoo paper

ModelEffortHarnessScoreCost per taskOutput tokens
Qwen3-235B-A22B MT-TIR 61% 14,066
DeepSeek-R1 May 2025 MT-TIR 55.3% 11,284
Qwen3-235B-A22B PoT 52.8% 10,552
Qwen3-32B PoT 49.9% 9,893
Qwen3-32B MT-TIR 47.3% 11,899
Qwen3-235B-A22B No tools 45.8% 9,160
Qwen3-8B PoT 43% 10,855
Qwen3-32B No tools 42.4% 9,160
DeepSeek-R1 May 2025 No tools 41.9% 8,278
Qwen3-8B MT-TIR 41% 15,462
Qwen3-8B No tools 37.6% 11,942
DeepSeek-R1 May 2025 PoT 36.6% 8,310

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 Aug 2025

Source: The ReasonZoo 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%) Qwen3-235B-A22B 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Qwen3-235B-A22B

Source: Official ReasonZoo 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 18 months ago, in May 2025, for DeepSeek-R1. 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 (7 entries).
  • Pass Models scored 4 models we track, plus 3 other entries on its leaderboard.
  • Warning Newest score The newest came 18 months ago, in May 2025, for DeepSeek-R1.
  • Note Frontier models None of the top 10 has a score. The strongest tested is DeepSeek-R1 (#122).
  • Pass Headroom The best is 61%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 12 rows for models we track and 3 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