Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

HMMT 2025 leaderboard — the family board.

7 TRACKED VARIANTS · 4 FIRST-CLASS BOARDS

One family, 7 tracked prints: HMMT Feb. 2025, HMMT Feb. 2026 and HMMT Nov. 2025 are cited standalone and keep live boards of their own. Scoring configs and splits of the base board render as sections of this page, each under its own honest label.

● LIVE DATA60 MODELS ON THIS BOARDLEADER GPT-5.2 PRO AT 100

First-class boards in this familyCITED STANDALONE · EACH KEEPS ITS OWN LIVE BOARD

#BoardModels scored
01HMMT 2025THIS PAGE60

The board — HMMT 2025TOP 60 OF 60 MEASURED · BEST SCORE PER MODEL, DEFAULT VARIANT

#ModelVendorScoreVLast seen
01GPT-5.2 ProOpenAI1002026-08-23
02GPT-5.2OpenAI99.42026-08-23
03DeepSeek-V3.2-SpecialeDeepSeek99.22026-08-23
04Kimi K2.5Moonshot95.42026-08-23
05Nemotron 3 Super 120B A12BNVIDIA94.72026-08-23
06Qwen3.6 PlusAlibaba94.62026-08-23
07PaCoRe-8BStepFun94.52026-06-25
08GLM-5.2Z.ai94.42026-08-23
09GLM-5.1Z.ai942026-08-23
10GPT-5OpenAI93.32026-08-23
11Grok 4 FastSpaceXAI93.32026-08-23
12Qwen3.5 397B A17BAlibaba92.72026-08-23
13Step3 VL 10BStepFun92.12026-06-05
14Qwen3.6 27BAlibaba90.72026-08-23
15Qwen3.5 122B A10BAlibaba90.32026-08-23
16DeepSeek-V3.2DeepSeek90.22026-09-24
17Grok 4SpaceXAI902026-08-23
18Qwen3.5 27BAlibaba89.82026-08-23
19Kimi K2 ThinkingMoonshot89.42026-06-25
20Qwen3.5 35B A3BAlibaba89.22026-08-23
21Granite 4.2 30BIBM89.22026-08-31
22Qwen3.6 35B A3BAlibaba89.12026-08-23
23GLM-4.6Z.ai88.72026-06-25
24GPT-5 miniOpenAI87.82026-08-23
25DeepSeek-V3.1-TerminusDeepSeek86.12026-06-25
26Sarvam 105BSarvam AI85.82026-08-23
27MiMo V2 FlashXiaomi84.42026-08-23
28DeepSeek-V3.2-ExpDeepSeek83.62026-08-23
29Qwen3.5 9BAlibaba82.92026-08-23
30Qwen3 235B ThinkingAlibaba82.32026-06-25
31Granite 4.2 8BIBM78.32026-08-31
32Qwen3 VL 235B A22B ReasoningAlibaba77.42026-08-23
33Qwen3.5 4BAlibaba76.82026-08-23
34GPT-5 nanoOpenAI75.62026-08-23
35RLVR-8BStepFun75.42026-06-25
36Claude Sonnet 4.5Anthropic74.62026-05-15
37Sarvam 30BSarvam AI74.22026-08-23
38Grok 3 Mini ReasoningSpaceXAI742026-07-13
39Qwen3 Next 80B A3BAlibaba73.92026-08-23
40Qwen3 VL 30B A3B ReasoningAlibaba67.62026-08-23
41Granite 4.2 3BIBM66.72026-08-31
42Gemini 2.5 ProGoogle65.72026-06-05
43Qwen3 VL Thinking (8B)Alibaba60.62026-08-23
44Qwen3 VL 235B A22B InstructAlibaba57.42026-08-23
45GLM-4.6V (106B-A12B)Z.ai57.32026-06-05
46Qwen3 235B A22B Instruct 2507Alibaba55.42026-08-23
47Qwen3 Next 80B A3B InstructAlibaba54.12026-08-23
48Qwen3 VL 4B (Reasoning)Alibaba53.12026-08-23
49Seed 1.5 VLByteDance51.32026-06-05
50Qwen3 VL 30B A3B InstructAlibaba50.62026-08-23
51Kimi K2 InstructMoonshot38.82026-08-23
52GPT-4.1 miniOpenAI352026-08-23
53DeepSeek-V3.1DeepSeek33.52026-08-23
54Qwen3 VL 8B InstructAlibaba32.52026-08-23
55Qwen3 VL 4B InstructAlibaba30.72026-08-23
56GPT-4.1OpenAI28.92026-08-23
57DeepSeek-V3DeepSeek27.52026-06-15
58Claude Opus 4Anthropic15.92026-06-15
59Claude Sonnet 4Anthropic15.92026-06-15
60Qwen3 235B A22BAlibaba11.92026-06-15
◆ IV — INDEPENDENTLY VERIFIED · ◆ AA — AGGREGATOR · ◆ VA — VENDOR · ◇ SA — SOURCE-ATTRIBUTED

Variants of this boardSCORING CONFIGS & SPLITS — NOT CITED STANDALONE, SO THEY LIVE HERE

Other tracked prints

THIN BOARDSunder 5 published models each — listed, never hidden
PrintLeaderScoreModels
HMMT Feb25Granite 4.2 30B89.23
HMMT Feb25 (no tools)Nemotron 3 Super 120B A12B94.22
HMMT Feb25 (with tools)Nemotron 3 Super 120B A12B95.52
THE FAMILY BOARD · EVERY PRINT LABELED · EVERY SCORE SOURCE-ATTRIBUTED ON ITS OWN LIVE BOARDAll leaderboardsThe frontier boardModels by vendor