Benchmark Almanac

VerifierBench

CompassVerifier-32B 90.8% tops its leaderboard GPT-4.1 81.1% leads the models we track
Not in model cards

Judge models decide whether proposed answers match references in math, science, and other subjects.

Released
August 2025
Kind
Other
Level
University
Scored
Checked automatically
Answers
Prediction
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Generated
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Qwen3-235B-A22B #131 on the capability index
Last model added
April 2025
Scores refreshed
November 20th, 2026
19
Entries in its paper, 9 models we track

CompassVerifier-32B tops the leaderboard with 90.8% GPT-4.1 leads with 81.1% among models released by April 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. 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 GPT-4.1, at 81.1%. This view shows only models we track, each at its best.

The newest model with a score here came out in April 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 CompassVerifier-32B 90.8%
  2. 2 CompassVerifier-7B 87.5%
  3. 3 CompassVerifier-3B 85%
  4. 4 GPT-4.1 81.1%
  5. 5 xVerify-8B-I 80.8%
  6. 6 xVerify-0.5B-I 79.8%
  7. 7 DeepSeek-V3 Mar 2025 79.1%
  8. 8 xVerify-9B-C 79%
  9. 9 Tencent-Qwen2.5-7B-RLVR 78.5%
  10. 10 GPT-4o Aug 2024 76.4%
  11. 11 Qwen3-14B 73.3%
  12. 12 Qwen3-235B-A22B 71%
  13. 13 Qwen3-32B 70.7%
  14. 14 Qwen2.5-72B 66.6%
  15. 15 Qwen2.5-32B-Instruct 63.8%
  16. 16 Qwen3-30B-A3B 61%
  17. 17 Qwen3-8B 59.1%
  18. 18 Qwen2.5-7B-Instruct 57.9%
  19. 19 Qwen2.5-14B-Instruct 57.4%

Source: The VerifierBench paper

#EntryScoreSettingDateSource
1 CompassVerifier-32B 90.8% Official leaderboard
2 CompassVerifier-7B 87.5% Official leaderboard
3 CompassVerifier-3B 85% Official leaderboard
4 GPT-4.1 81.1% Apr 2025 Official leaderboard
5 xVerify-8B-I 80.8% Official leaderboard
6 xVerify-0.5B-I 79.8% Official leaderboard
7 DeepSeek-V3 Mar 2025 79.1% Mar 2025 Official leaderboard
8 xVerify-9B-C 79% Official leaderboard
9 Tencent-Qwen2.5-7B-RLVR 78.5% Official leaderboard
10 GPT-4o Aug 2024 76.4% Aug 2024 Official leaderboard
11 Qwen3-14B 73.3% Apr 2025 Official leaderboard
12 Qwen3-235B-A22B 71% Apr 2025 Official leaderboard
13 Qwen3-32B 70.7% Apr 2025 Official leaderboard
14 Qwen2.5-72B 66.6% Sep 2024 Official leaderboard
15 Qwen2.5-32B-Instruct 63.8% Official leaderboard
16 Qwen3-30B-A3B 61% Apr 2025 Official leaderboard
17 Qwen3-8B 59.1% Apr 2025 Official leaderboard
18 Qwen2.5-7B-Instruct 57.9% Official leaderboard
19 Qwen2.5-14B-Instruct 57.4% Official leaderboard
#ModelScoreSettingDateSource
1 GPT-4.1 81.1% Apr 2025 Official leaderboard
2 DeepSeek-V3 Mar 2025 79.1% Mar 2025 Official leaderboard
3 GPT-4o Aug 2024 76.4% Aug 2024 Official leaderboard
4 Qwen3-14B 73.3% Apr 2025 Official leaderboard
5 Qwen3-235B-A22B 71% Apr 2025 Official leaderboard
6 Qwen3-32B 70.7% Apr 2025 Official leaderboard
7 Qwen2.5-72B 66.6% Sep 2024 Official leaderboard
8 Qwen3-30B-A3B 61% Apr 2025 Official leaderboard
9 Qwen3-8B 59.1% Apr 2025 Official leaderboard

Number one over time

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

OpenAI
Benchmark paper Not tested since Aug 2025
GPT-4.1 81.1%

Source: The VerifierBench 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 Best so far
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) GPT-4.1 GPT-4o 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) GPT-4.1 GPT-4o

Source: Official VerifierBench 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 Other

All 69 in Other
  • Artificial Analysis Intelligence Index v5 In use Version 5 of Artificial Analysis's index: nine evaluations in four equal categories, most with private test sets, run three times each in its own agent harness. Claude Fable 5.2 64.6
  • Personal Superintelligence Index (Meta) In use Meta's composite of seven personal and knowledge-work evals: Agents' Last Exam, AutomationBench-AA, FocusBench, JobBench, GDP.pdf, VBT and DRACO.
  • Vals Science Index In use Agents are scored on the equal average of BioMysteryBench, MysteryMechanism and Terminal-Bench-Science. GPT-6.2 Astra 68.4%
  • Decision Index Public v0.3 In use Decision models return probabilities over labeled options in public tasks covering knowledge, language, retrieval, tools, and arts. A chance-corrected weighted aggregate combines the task scores. Granite Switch 4.1 30B 51.3
  • SAGE v1.1 (Vals) In use Models grade handwritten undergraduate math and computer science work against rubrics. Scored by class-balanced accuracy of each rubric check against expert grading. Claude Opus 5.5 91.3%
  • MTEB Fading Massive Text Embedding Benchmark: retrieval, clustering, classification across languages and modalities. Qwen3-Embedding-8B 72.2%

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 19 months ago, in April 2025, for Qwen3-14B. 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 (19 entries).
  • Pass Models scored 9 models we track, plus 10 other entries on its leaderboard.
  • Warning Newest score The newest came 19 months ago, in April 2025, for Qwen3-14B.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Qwen3-235B-A22B (#131).
  • Pass Headroom The best is 81.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 9 rows for models we track and 10 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