Benchmark Almanac

FINESSE-Bench

GPT-5.5 0.90 leads since May 2026
Not in model cards

Models answer financial certification questions and interpret trading indicators. The benchmark checks numerical and multiple-choice answers and uses a model judge for free responses.

Released
May 2026
Kind
Finance
Level
University
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English, Russian
In and out
Text → Text
Interaction
Single turn
Items from
Exams and contests
Refresh
One fixed set
Run by
Anyone
In views
Law and finance
Strongest model tested
GPT-5.5 #21 on the capability index
Last model added
June 2026
Scores refreshed
November 20th, 2026
45
Entries on its own leaderboard, 36 models we track

GPT-5.5 leads with 0.90 among models released by June 2026 GPT-5.5 leads with 0.90 among models released by June 2026

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. 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. This view shows only models we track, each at its best.

The newest model with a score here came out in June 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GPT-5.5 0.90
  2. 2 Claude Opus 4.8 0.88
  3. 3 Kimi K2.5 0.87
  4. 4 GPT-5 0.87
  5. 5 Qwen3.5-Plus 0.87
  6. 6 Claude Sonnet 4.6 0.87
  7. 7 GPT-5.2 0.87
  8. 8 Qwen3.5-122B-A17B 0.85
  9. 9 Claude Sonnet 5 0.85
  10. 10 GLM-5 0.85
  11. 11 Qwen3.5 397B-A17B 0.85
  12. 12 GLM-4.7 0.85
  13. 13 Qwen3.5-27B 0.84
  14. 14 Qwen3.5-Flash 0.84
  15. 15 Qwen3-235B-A22B-Thinking Jul 2025 0.84
  16. 16 Qwen3.5-35B-A3B 0.83
  17. 17 Claude 3.7 Sonnet 0.83
  18. 18 DeepSeek-R1 May 2025 0.82
  19. 19 GLM-4.5 0.82
  20. 20 MiniMax-M2.5 0.82
  21. 21 GPT-5.4 0.81
  22. 22 Qwen3-32B think 0.80
  23. 23 GLM-4.6 0.80
  24. 24 Llama 4 Maverick 0.79
  25. 25 T-pro-it-2.0 0.79
  26. 26 MiniMax-M1-80k 0.79
  27. 27 Qwen3.5-9B 0.79
  28. 28 MiniMax-M2.1 0.79
  29. 29 GLM-4.5-Air 0.79
  30. 30 DeepSeek-V3.2 0.79
  31. 31 GPT-4.1 0.78
  32. 32 Qwen3-14B think 0.77
  33. 33 Llama 4 Scout 0.75
  34. 34 Qwen3-8B think 0.74
  35. 35 Llama 3.3 70B 0.72
  36. 36 GigaChat-2-Max 0.71
  37. 37 YandexGPT 5.1 Pro 0.71
  38. 38 GigaChat3-702B-A36B-fp8 0.70
  39. 39 MiMo-V2-Flash 0.69
  40. 40 Mistral Small 3.2 0.67
  41. 41 Llama 3.1-70B 0.66
  42. 42 Fin-o1-8B think 0.66
  43. 43 Fin-R1 think 0.65
  44. 44 GigaChat3-10B-A1.8B-bf16 0.59
  45. 45 Fino1-8B think 0.57

Source: Official FINESSE-Bench leaderboard

Source: Official FINESSE-Bench leaderboard

#EntryScoreSettingDateSource
1 GPT-5.5 0.90 Apr 2026 Official leaderboard
2 Claude Opus 4.8 0.88 May 2026 Official leaderboard
3 Kimi K2.5 0.87 Jan 2026 Official leaderboard
4 GPT-5 0.87 Aug 2025 Official leaderboard
5 Qwen3.5-Plus 0.87 Feb 2026 Official leaderboard
6 Claude Sonnet 4.6 0.87 Feb 2026 Official leaderboard
7 GPT-5.2 0.87 Dec 2025 Official leaderboard
8 Qwen3.5-122B-A17B 0.85 Official leaderboard
9 Claude Sonnet 5 0.85 Jun 2026 Official leaderboard
10 GLM-5 0.85 Feb 2026 Official leaderboard
11 Qwen3.5 397B-A17B 0.85 Feb 2026 Official leaderboard
12 GLM-4.7 0.85 Dec 2025 Official leaderboard
13 Qwen3.5-27B 0.84 Feb 2026 Official leaderboard
14 Qwen3.5-Flash 0.84 Feb 2026 Official leaderboard
15 Qwen3-235B-A22B-Thinking Jul 2025 0.84 Jul 2025 Official leaderboard
16 Qwen3.5-35B-A3B 0.83 Feb 2026 Official leaderboard
17 Claude 3.7 Sonnet 0.83 Feb 2025 Official leaderboard
18 DeepSeek-R1 May 2025 0.82 May 2025 Official leaderboard
19 GLM-4.5 0.82 Aug 2025 Official leaderboard
20 MiniMax-M2.5 0.82 Feb 2026 Official leaderboard
21 GPT-5.4 0.81 Mar 2026 Official leaderboard
22 Qwen3-32B 0.80 think Apr 2025 Official leaderboard
23 GLM-4.6 0.80 Sep 2025 Official leaderboard
24 Llama 4 Maverick 0.79 Apr 2025 Official leaderboard
25 T-pro-it-2.0 0.79 Official leaderboard
26 MiniMax-M1-80k 0.79 Jun 2025 Official leaderboard
27 Qwen3.5-9B 0.79 Feb 2026 Official leaderboard
28 MiniMax-M2.1 0.79 Dec 2025 Official leaderboard
29 GLM-4.5-Air 0.79 Jul 2025 Official leaderboard
30 DeepSeek-V3.2 0.79 Dec 2025 Official leaderboard
31 GPT-4.1 0.78 Apr 2025 Official leaderboard
32 Qwen3-14B 0.77 think Apr 2025 Official leaderboard
33 Llama 4 Scout 0.75 Apr 2025 Official leaderboard
34 Qwen3-8B 0.74 think Apr 2025 Official leaderboard
35 Llama 3.3 70B 0.72 Dec 2024 Official leaderboard
36 GigaChat-2-Max 0.71 Official leaderboard
37 YandexGPT 5.1 Pro 0.71 Official leaderboard
38 GigaChat3-702B-A36B-fp8 0.70 Official leaderboard
39 MiMo-V2-Flash 0.69 Dec 2025 Official leaderboard
40 Mistral Small 3.2 0.67 Jun 2025 Official leaderboard
41 Llama 3.1-70B 0.66 Jul 2024 Official leaderboard
42 Fin-o1-8B 0.66 think Official leaderboard
43 Fin-R1 0.65 think Official leaderboard
44 GigaChat3-10B-A1.8B-bf16 0.59 Official leaderboard
45 Fino1-8B 0.57 think Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5.5 0.90 Apr 2026 Official leaderboard
2 Claude Opus 4.8 0.88 May 2026 Official leaderboard
3 Kimi K2.5 0.87 Jan 2026 Official leaderboard
4 GPT-5 0.87 Aug 2025 Official leaderboard
5 Qwen3.5-Plus 0.87 Feb 2026 Official leaderboard
6 Claude Sonnet 4.6 0.87 Feb 2026 Official leaderboard
7 GPT-5.2 0.87 Dec 2025 Official leaderboard
8 Claude Sonnet 5 0.85 Jun 2026 Official leaderboard
9 GLM-5 0.85 Feb 2026 Official leaderboard
10 Qwen3.5 397B-A17B 0.85 Feb 2026 Official leaderboard
11 GLM-4.7 0.85 Dec 2025 Official leaderboard
12 Qwen3.5-27B 0.84 Feb 2026 Official leaderboard
13 Qwen3.5-Flash 0.84 Feb 2026 Official leaderboard
14 Qwen3-235B-A22B-Thinking Jul 2025 0.84 Jul 2025 Official leaderboard
15 Qwen3.5-35B-A3B 0.83 Feb 2026 Official leaderboard
16 Claude 3.7 Sonnet 0.83 Feb 2025 Official leaderboard
17 DeepSeek-R1 May 2025 0.82 May 2025 Official leaderboard
18 GLM-4.5 0.82 Aug 2025 Official leaderboard
19 MiniMax-M2.5 0.82 Feb 2026 Official leaderboard
20 GPT-5.4 0.81 Mar 2026 Official leaderboard
21 Qwen3-32B 0.80 think Apr 2025 Official leaderboard
22 GLM-4.6 0.80 Sep 2025 Official leaderboard
23 Llama 4 Maverick 0.79 Apr 2025 Official leaderboard
24 MiniMax-M1-80k 0.79 Jun 2025 Official leaderboard
25 Qwen3.5-9B 0.79 Feb 2026 Official leaderboard
26 MiniMax-M2.1 0.79 Dec 2025 Official leaderboard
27 GLM-4.5-Air 0.79 Jul 2025 Official leaderboard
28 DeepSeek-V3.2 0.79 Dec 2025 Official leaderboard
29 GPT-4.1 0.78 Apr 2025 Official leaderboard
30 Qwen3-14B 0.77 think Apr 2025 Official leaderboard
31 Llama 4 Scout 0.75 Apr 2025 Official leaderboard
32 Qwen3-8B 0.74 think Apr 2025 Official leaderboard
33 Llama 3.3 70B 0.72 Dec 2024 Official leaderboard
34 MiMo-V2-Flash 0.69 Dec 2025 Official leaderboard
35 Mistral Small 3.2 0.67 Jun 2025 Official leaderboard
36 Llama 3.1-70B 0.66 Jul 2024 Official leaderboard

Rankings by part

We read 14 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Public benchmarks, Exam-like and Trading / TA. GPT-5.5 leads 9 of them, and 4 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Public benchmarks

36 models we track have a score here.

  1. 1 Claude Opus 4.8 ↑1 0.91
  2. 2 GPT-5.5 ↓1 0.91
  3. 3 Claude Sonnet 4.6 ↑3 0.91
  4. 4 Claude Sonnet 5 ↑4 0.90
  5. 5 Kimi K2.5 ↓2 0.90
  6. 6 GPT-5.2 ↑1 0.90
  7. 7 MiniMax-M2.5 ↑12 0.89
  8. 8 GPT-5 ↓4 0.89
  9. 9 GPT-5.4 ↑11 0.89
  10. 10 Qwen3.5-Plus ↓5 0.89

Number one over time

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

OpenAI
Own leaderboard Not tested since Jun 2026
GPT-5.5 0.90

Source: Official FINESSE-Bench 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
0.70 0.80 0.90 2025 2026 0.70 0.80 0.90 2025 2026

Source: Official FINESSE-Bench 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 Law, finance and medicine

All 637 in Law, finance and medicine
  • MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy. Qwen 3.8 Max 80.4%
  • HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
  • HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
  • FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy. Ling-3.1-flash 81.7%
  • KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
  • Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (45 entries).
  • Pass Models scored 36 models we track, plus 9 other entries on its leaderboard.
  • Pass Newest score The newest is from June 2026, for Claude Sonnet 5.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on limexailab.github.io: 36 rows for models we track and 9 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