Benchmark Almanac

HMMT Nov 2025

GPT-5.2 99.2% leads since December 2025
GPT-5.2 99.2% leads since December 2025
Fading

Models solve the 30 problems of the November 2025 Harvard-MIT Mathematics Tournament. Final answers are checked automatically against the answer key.

Released
November 2025
Kind
Competition problems with answers
Level
Competition
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Strongest model tested
GPT-5.2 #50 on the capability index
Last in a launch post
June 2026, dropped after it saturated
Last model added
June 2026
Scores refreshed
November 12th, 2026
22
Models scored on its own leaderboard
24
Launch posts list it, from 9 labs
11
System cards report it, from 9 labs

GPT-5.2 leads with 99.2% among models released by February 2026

Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.

GPT-5.2 leads with 99.2%, claims included

Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.

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

22 models. Scroll sideways for the rest.

  1. 99.2 GPT-5.2
  2. 94.2 Step 3.5 Flash
  3. 94.2 GLM-5
  4. 93.3 Gemini 3 Pro
  5. 93.3 Grok 4.1 Fast
  6. 93.3 DeepSeek-V3.2-Speciale
  7. 93.3 Gemini 3 Flash
  8. 91.7 GLM-4.6
  9. 91.7 GPT-5.1
  10. 90.8 Grok 4 Fast
  11. 90.0 gpt-oss-120b
  12. 90.0 DeepSeek-V3.2
  13. 89.2 GPT-5
  14. 89.2 Kimi K2 Thinking
  15. 89.2 Kimi K2.5
  16. 88.3 Grok 4
  17. 84.2 GPT-5 mini
  18. 84.2 DeepSeek-V3.2-Exp
  19. 81.7 GPT-5 nano
  20. 80.0 Gemini 2.5 Pro
  21. 79.2 Falcon H1R 7B
  22. 75.0 QED-Nano
  1. 1 GPT-5.2 xhigh 99.2%
  2. 2 Step 3.5 Flash 94.2%
  3. 3 GLM-5 94.2%
  4. 4 Gemini 3 Pro 93.3%
  5. 5 Grok 4.1 Fast reasoning 93.3%
  6. 6 DeepSeek-V3.2-Speciale 93.3%
  7. 7 Gemini 3 Flash high 93.3%
  8. 8 GLM-4.6 91.7%
  9. 9 GPT-5.1 high 91.7%
  10. 10 Grok 4 Fast 90.8%
  11. 11 gpt-oss-120b high 90%
  12. 12 DeepSeek-V3.2 thinking 90%
  13. 13 GPT-5 high 89.2%
  14. 14 Kimi K2 Thinking 89.2%
  15. 15 Kimi K2.5 high 89.2%
  16. 16 Grok 4 88.3%
  17. 17 GPT-5 mini high 84.2%
  18. 18 DeepSeek-V3.2-Exp thinking 84.2%
  19. 19 GPT-5 nano high 81.7%
  20. 20 Gemini 2.5 Pro Jun 2025 80%
  21. 21 Falcon H1R 7B 79.2%
  22. 22 QED-Nano 75%

Source: Official HMMT Nov 2025 leaderboard

45 models. Scroll sideways for the rest.

  1. 99.2 GPT-5.2
  2. 98.3 Ring-2.5-1T
  3. 96.9 GLM-5
  4. 94.7 Qwen3-Max
  5. 94.6 Qwen 3.6 Plus
  6. 94.4 DeepSeek-V3.2-Speciale
  7. 94.4 GLM-5.2
  8. 94.2 Step 3.5 Flash
  9. 94.0 GLM-5.1
  10. 93.5 GLM-4.7
  11. 93.3 Gemini 3 Pro
  12. 93.3 Grok 4.1 Fast
  13. 93.3 Gemini 3 Flash
  14. 93.3 Seed2.0 Pro
  15. 92.7 Qwen3.5 397B-A17B
  16. 91.7 GLM-4.6
  17. 91.7 GPT-5.1
  18. 91.1 Kimi K2.5
  19. 90.8 Grok 4 Fast
  20. 90.7 Qwen3.6 27B
  21. 90.3 Qwen3.5-122B-A10B
  22. 90.2 DeepSeek-V3.2
  23. 90.0 gpt-oss-120b
  24. 89.8 Qwen3.5-27B
  25. 89.2 Qwen3.5-35B-A3B
  26. 89.2 GPT-5
  27. 89.2 Kimi K2 Thinking
  28. 89.1 Qwen 3.6 35B-A3B
  29. 88.3 Grok 4
  30. 86.8 K-EXAONE
  31. 85.8 Sarvam 105B
  32. 84.4 Qwen3.5-Omni-Plus
  33. 84.2 GPT-5 mini
  34. 84.2 DeepSeek-V3.2-Exp
  35. 82.9 Qwen3.5-9B
  36. 81.9 Ling-2.6-1T
  37. 81.7 GPT-5 nano
  38. 80.0 Gemini 2.5 Pro
  39. 80.0 Solar Open 100B
  40. 79.2 Falcon H1R 7B
  41. 76.8 Qwen3.5-4B
  42. 75.6 Qwen3-Coder-Next
  43. 75.0 QED-Nano
  44. 74.2 Sarvam 30B
  45. 19.6 Qwen3.5-2B
  1. 1 GPT-5.2 xhigh 99.2%
  2. 2 Ring-2.5-1T 98.3%
  3. 3 GLM-5 96.9%
  4. 4 Qwen3-Max 94.7%
  5. 5 Qwen 3.6 Plus 94.6%
  6. 6 DeepSeek-V3.2-Speciale 94.4%
  7. 7 GLM-5.2 94.4%
  8. 8 Step 3.5 Flash 94.2%
  9. 9 GLM-5.1 94%
  10. 10 GLM-4.7 93.5%
  11. 11 Gemini 3 Pro 93.3%
  12. 12 Grok 4.1 Fast reasoning 93.3%
  13. 13 Gemini 3 Flash high 93.3%
  14. 14 Seed2.0 Pro 93.3%
  15. 15 Qwen3.5 397B-A17B 92.7%
  16. 16 GLM-4.6 91.7%
  17. 17 GPT-5.1 high 91.7%
  18. 18 Kimi K2.5 91.1%
  19. 19 Grok 4 Fast 90.8%
  20. 20 Qwen3.6 27B 90.7%
  21. 21 Qwen3.5-122B-A10B 90.3%
  22. 22 DeepSeek-V3.2 90.2%
  23. 23 gpt-oss-120b high 90%
  24. 24 Qwen3.5-27B 89.8%
  25. 25 Qwen3.5-35B-A3B 89.2%
  26. 26 GPT-5 high 89.2%
  27. 27 Kimi K2 Thinking 89.2%
  28. 28 Qwen 3.6 35B-A3B 89.1%
  29. 29 Grok 4 88.3%
  30. 30 K-EXAONE 86.8%
  31. 31 Sarvam 105B 85.8%
  32. 32 Qwen3.5-Omni-Plus 84.4%
  33. 33 GPT-5 mini high 84.2%
  34. 34 DeepSeek-V3.2-Exp thinking 84.2%
  35. 35 Qwen3.5-9B 82.9%
  36. 36 Ling-2.6-1T 81.9%
  37. 37 GPT-5 nano high 81.7%
  38. 38 Gemini 2.5 Pro Jun 2025 80%
  39. 39 Solar Open 100B 80%
  40. 40 Falcon H1R 7B 79.2%
  41. 41 Qwen3.5-4B 76.8%
  42. 42 Qwen3-Coder-Next 75.6%
  43. 43 QED-Nano 75%
  44. 44 Sarvam 30B 74.2%
  45. 45 Qwen3.5-2B 19.6%

Sources: Official HMMT Nov 2025 leaderboard and labs’ model cards

#ModelScoreSettingDateSource
1 GPT-5.2 99.2% xhigh Dec 2025 Official leaderboard
2 Step 3.5 Flash 94.2% Feb 2026 Official leaderboard
3 GLM-5 94.2% Feb 2026 Official leaderboard
4 Gemini 3 Pro 93.3% Nov 2025 Official leaderboard
5 Grok 4.1 Fast 93.3% reasoning Nov 2025 Official leaderboard
6 DeepSeek-V3.2-Speciale 93.3% Dec 2025 Official leaderboard
7 Gemini 3 Flash 93.3% high Dec 2025 Official leaderboard
8 GLM-4.6 91.7% Sep 2025 Official leaderboard
9 GPT-5.1 91.7% high Nov 2025 Official leaderboard
10 Grok 4 Fast 90.8% Sep 2025 Official leaderboard
11 gpt-oss-120b 90% high Aug 2025 Official leaderboard
12 DeepSeek-V3.2 90% thinking Dec 2025 Official leaderboard
13 GPT-5 89.2% high Aug 2025 Official leaderboard
14 Kimi K2 Thinking 89.2% Nov 2025 Official leaderboard
15 Kimi K2.5 89.2% high Jan 2026 Official leaderboard
16 Grok 4 88.3% Jul 2025 Official leaderboard
17 GPT-5 mini 84.2% high Aug 2025 Official leaderboard
18 DeepSeek-V3.2-Exp 84.2% thinking Sep 2025 Official leaderboard
19 GPT-5 nano 81.7% high Aug 2025 Official leaderboard
20 Gemini 2.5 Pro Jun 2025 80% Jun 2025 Official leaderboard
21 Falcon H1R 7B 79.2% Jan 2026 Official leaderboard
22 QED-Nano 75% Feb 2026 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5.2 99.2% xhigh Dec 2025 Official leaderboard
2 Ring-2.5-1T 98.3% Feb 2026 Lab model card
3 GLM-5 96.9% Feb 2026 Lab model card
4 Qwen3-Max 94.7% Jan 2026 Lab model card
5 Qwen 3.6 Plus 94.6% Apr 2026 Lab model card
6 DeepSeek-V3.2-Speciale 94.4% Dec 2025 Lab model card
7 GLM-5.2 94.4% Jun 2026 Lab model card
8 Step 3.5 Flash 94.2% Feb 2026 Official leaderboard
9 GLM-5.1 94% Apr 2026 Lab model card
10 GLM-4.7 93.5% Dec 2025 Lab model card
11 Gemini 3 Pro 93.3% Nov 2025 Official leaderboard
12 Grok 4.1 Fast 93.3% reasoning Nov 2025 Official leaderboard
13 Gemini 3 Flash 93.3% high Dec 2025 Official leaderboard
14 Seed2.0 Pro 93.3% Feb 2026 Lab model card
15 Qwen3.5 397B-A17B 92.7% Feb 2026 Lab model card
16 GLM-4.6 91.7% Sep 2025 Official leaderboard
17 GPT-5.1 91.7% high Nov 2025 Official leaderboard
18 Kimi K2.5 91.1% Jan 2026 Lab model card
19 Grok 4 Fast 90.8% Sep 2025 Official leaderboard
20 Qwen3.6 27B 90.7% Apr 2026 Lab model card
21 Qwen3.5-122B-A10B 90.3% Feb 2026 Lab model card
22 DeepSeek-V3.2 90.2% Dec 2025 Lab model card
23 gpt-oss-120b 90% high Aug 2025 Official leaderboard
24 Qwen3.5-27B 89.8% Feb 2026 Lab model card
25 Qwen3.5-35B-A3B 89.2% Feb 2026 Lab model card
26 GPT-5 89.2% high Aug 2025 Official leaderboard
27 Kimi K2 Thinking 89.2% Nov 2025 Official leaderboard
28 Qwen 3.6 35B-A3B 89.1% Apr 2026 Lab model card
29 Grok 4 88.3% Jul 2025 Official leaderboard
30 K-EXAONE 86.8% Dec 2025 Lab model card
31 Sarvam 105B 85.8% Mar 2026 Lab model card
32 Qwen3.5-Omni-Plus 84.4% Mar 2026 Lab model card
33 GPT-5 mini 84.2% high Aug 2025 Official leaderboard
34 DeepSeek-V3.2-Exp 84.2% thinking Sep 2025 Official leaderboard
35 Qwen3.5-9B 82.9% Mar 2026 Lab model card
36 Ling-2.6-1T 81.9% Jun 2026 Lab model card
37 GPT-5 nano 81.7% high Aug 2025 Official leaderboard
38 Gemini 2.5 Pro Jun 2025 80% Jun 2025 Official leaderboard
39 Solar Open 100B 80% Dec 2025 Lab model card
40 Falcon H1R 7B 79.2% Jan 2026 Official leaderboard
41 Qwen3.5-4B 76.8% Mar 2026 Lab model card
42 Qwen3-Coder-Next 75.6% Feb 2026 Lab model card
43 QED-Nano 75% Feb 2026 Official leaderboard
44 Sarvam 30B 74.2% Mar 2026 Lab model card
45 Qwen3.5-2B 19.6% Mar 2026 Lab model card

Labs report scores for 28 models in their cards. Switch the scores above to rank them together.

The best score, 99.2%, costs $0.19 a task

Runs on the cost frontier score higher than any run that's cheaper. Step 3.5 Flash comes within 5 points of the best for $0.01 a task, 14 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

xAI OpenAI StepFun Cost frontier One model at each effort level
0% 20% 40% 60% 80% 100% $0.001 $0.01 $0.10 $1.00 Cost per task, log scale GPT-5.2, xhigh Step 3.5 Flash Grok 4.1 Fast, reasoning 0% 20% 40% 60% 80% 100% $0.001 $0.01 $0.10 $1.00 Cost per task, log scale GPT-5.2, xhigh Step 3.5 Flash

Source: Official HMMT Nov 2025 leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
GPT-5.2 xhigh 99.2% $0.19 13,192
GPT-5.2 high 95.8% $0.14 10,015
Step 3.5 Flash 94.2% $0.01 45,001
GLM-5 94.2% $0.10 30,083
Gemini 3 Pro 93.3% $0.18 14,837
Grok 4.1 Fast reasoning 93.3% $0.005 10,401
DeepSeek-V3.2-Speciale 93.3% $0.01 25,249
Gemini 3 Flash high 93.3% $0.06 20,413
GLM-4.6 91.7% $0.04 16,843
GPT-5.1 high 91.7% $0.20 19,593
Grok 4 Fast 90.8% $0.005 10,639
gpt-oss-120b high 90% $0.01 18,543
DeepSeek-V3.2 thinking 90% $0.007 17,165
GPT-5 high 89.2% $0.15 15,483
Kimi K2 Thinking 89.2% $0.07 28,752
Kimi K2.5 high 89.2% $0.08 26,085
Grok 4 88.3% $0.22 14,792
GPT-5 mini high 84.2% $0.03 14,859
DeepSeek-V3.2-Exp thinking 84.2% $0.007 17,348
GPT-5 nano high 81.7% $0.01 26,955
Gemini 2.5 Pro Jun 2025 80% $0.22 21,190
Falcon H1R 7B 79.2% $0.006 28,381
QED-Nano 75% 39,270

3 models have held number one on the official leaderboard since its release in November 2025 4 models have held number one with claims included since its release in November 2025

Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for the benchmark’s own leaderboard and what labs claim in their cards.

Z.ai Google DeepMind OpenAI DeepSeek Qwen Ant Group
Own leaderboard Passed 85% in Nov 2025, by GLM-4.6
GPT-5.2 99.2%
Lab claims Passed 85% in Dec 2025, by DeepSeek-V3.2-Speciale
Ring-2.5-1T 98.3%

Sources: Official HMMT Nov 2025 leaderboard and labs’ model cards

Grok 4 on the official leaderboard passed 85% in July 2025

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-5.2 DeepSeek-V3.2-Speciale gpt-oss-120b Gemini 2.5 Pro 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-5.2 GLM-4.6

Sources: Official HMMT Nov 2025 leaderboard and labs’ model cards

Labs headline it less often now, last in June 2026

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

DeepSeek
Z.ai
LG AI Research
Upstage
Qwen
StepFun
Ant Group
ByteDance Seed
Sarvam AI
DateLaunch postScoreAs the post writes it
Jun 16, 2026 GLM-5.2 94.4 HMMT Nov. 2025
Apr 22, 2026 Qwen3.6 27B 90.7 HMMT Nov 25
Apr 15, 2026 Qwen 3.6 35B-A3B 89.1 HMMT Nov 25
Apr 7, 2026 GLM-5.1 94 HMMT Nov. 2025
Apr 2, 2026 Qwen 3.6 Plus 94.6 HMMT Nov 25
Mar 30, 2026 Qwen3.5-Omni-Plus 84.4 HMMT Nov 25
Mar 6, 2026 Sarvam 30B 74.2 HMMT Nov 2025
Mar 6, 2026 Sarvam 105B 85.8 HMMT (Nov 25)
Mar 2, 2026 Qwen3.5-2B 19.6 HMMT Nov 25
Mar 2, 2026 Qwen3.5-4B 76.8 HMMT Nov 25
Mar 2, 2026 Qwen3.5-9B 82.9 HMMT Nov 25
Feb 24, 2026 Qwen3.5-35B-A3B 89.2 HMMT Nov 25
Feb 24, 2026 Qwen3.5-27B 89.8 HMMT Nov 25
Feb 24, 2026 Qwen3.5-122B-A10B 90.3 HMMT Nov 25
Feb 16, 2026 Qwen3.5 397B-A17B 92.7 HMMT Nov 25
Feb 14, 2026 Seed2.0 Pro 93.3 HMMT Nov 2025
Feb 11, 2026 GLM-5 96.9 HMMT Nov. 2025
Feb 10, 2026 Ring-2.5-1T 98.3 Best of HMMT (Nov. 25) (Mean@64) 97.34 / Heavy Thinking 98.33
Feb 2, 2026 Step 3.5 Flash 94 HMMT 2025 (Nov.)
Jan 23, 2026 Qwen3-Max 94.7 HMMT Nov 25
Dec 31, 2025 Solar Open 100B 80 HMMT 2025 (Nov)
Dec 31, 2025 K-EXAONE 86.8 HMMT Nov 2025
Dec 22, 2025 GLM-4.7 93.5 HMMT Nov. 2025
Dec 1, 2025 DeepSeek-V3.2 90.2 HMMT Nov 2025 (Pass@1)

11 system cards report it, from 9 labs

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Jun 13, 2026 Ling-2.6-1T technical report 81.9 HMMT-Nov25, Mean@64, CoT (Table 5) Capabilities
Apr 17, 2026 Qwen3.5-Omni-Plus technical report 84.4 HMMT Nov 25 (Table 4) Capabilities
Feb 28, 2026 Qwen3-Coder-Next technical report 75.6 HMMT25 Nov (Table 9) Capabilities
Feb 17, 2026 GLM-5 technical report 96.9 HMMT Nov. 2025 (Table 7) Capabilities
Feb 14, 2026 Seed2.0 Pro model card 93.3 HMMT Nov 2025 (Table 3) Capabilities
Feb 11, 2026 Step 3.5 Flash technical report 94 HMMT 2025 Nov., avg of 64 (Table 5). PaCoRe 97.8, with Python 98.0 (Table 12) Capabilities
Jan 30, 2026 Kimi K2.5 technical report 91.1 HMMT Nov 2025, average 24k output tokens (Table 5) Capabilities
Jan 11, 2026 Solar Open 100B technical report 80 HMMT 2025 (Nov) (Table 6) Capabilities
Jan 5, 2026 K-EXAONE technical report 86.8 HMMT Nov 2025 (Table 3, Reasoning mode). Non-reasoning 43.2 (Table 4) Capabilities
Dec 1, 2025 DeepSeek-V3.2 technical report 90.2 HMMT Nov 2025 (Pass@1), DeepSeek-V3.2 Thinking (Table 2) Capabilities
Dec 1, 2025 DeepSeek-V3.2-Speciale technical report 94.4 HMMT Nov 2025 (Pass@1), DeepSeek-V3.2-Speciale (Table 3) Capabilities

The lab's number comes out ahead in 4 of 5 pairs that differ

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. Every benchmark.

Model Lab says Board says Gap
GLM-5 96.9% HMMT Nov. 2025 94.2% Leaderboard +2.7 pts
Kimi K2.5 91.1% HMMT Nov 2025, average 24k output tokens (Table 5) 89.2% Leaderboard, high +1.9 pts
DeepSeek-V3.2-Speciale 94.4% HMMT Nov 2025 (Pass@1), DeepSeek-V3.2-Speciale (Table 3) 93.3% Leaderboard +1.1 pts
DeepSeek-V3.2 90.2% HMMT Nov 2025 (Pass@1) 90% Leaderboard, thinking +0.2 pts
Step 3.5 Flash 94% HMMT 2025 (Nov.) 94.2% Leaderboard −0.2 pts

More in Math

All 113 in Math
  • AIME 2026 In use Models solve the 30 integer-answer problems from the 2026 AIME I and II math competitions. Scored by exact answer match. GPT-5.5 100%
  • IMO-AnswerBench In use 400 olympiad problems with short verifiable answers, robustified against memorization. Scored by answer match. GLM-5.2 91%
  • HMMT 2026 In use Models solve the 33 final-answer problems from the February 2026 Harvard-MIT Mathematics Tournament. Scored by exact answer match. GPT-5.5 98.5%
  • PolyMath In use Models solve 9,000 parallel math problems in 18 languages across four difficulty levels, from school math to frontier math. Final answers are checked. Qwen3.7-Max 86.5%
  • MathArena (Apex, IMO, USAMO) In use Evaluates models on newly released math competitions after release. Final-answer or human/LLM-judged proof scores. Claude Opus 4.8 81.3%
  • MathArena Apex Shortlist In use Models solve 47 hard final-answer problems that MathArena selected from 2025 competitions, where the best models scored about 50% at selection. Scored by accuracy over repeated attempts. GPT-5.5 98.4%