Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

GPQA Diamond leaderboard — the family board.

4 TRACKED VARIANTS · 2 FIRST-CLASS BOARDS

One family, 4 tracked prints: GPQA (HF Open LLM) is cited standalone and keeps live board of its own. Scoring configs and splits of the base board render as sections of this page, each under its own honest label.

● LIVE DATA451 MODELS ON THIS BOARDLEADER GPT-6 ASTRA AT 96.3

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

#BoardModels scored
01GPQA DiamondTHIS PAGE451

The board — GPQA DiamondTOP 100 OF 451 MEASURED · BEST SCORE PER MODEL, DEFAULT VARIANT

#ModelVendorScoreVLast seen
01GPT-6 AstraOpenAI96.32026-09-04
02Gemini 3.8 FlashGoogle95.32026-09-03
03Grok 4.6SpaceXAI94.92026-08-22
04Claude Mythos PreviewAnthropic94.62026-08-23
05GPT-5.6 SolOpenAI94.62026-09-10
06Gemini 3.7 FlashGoogle94.52026-08-14
07GPT-5.4 ProOpenAI94.42026-06-21
08Gemini 3.1 ProGoogle94.32026-10-08
09Claude Opus 4.7Anthropic94.22026-10-04
10Muse Spark 1.3Meta94.12026-09-03
11Claude Fable 5.1Anthropic93.72026-10-04
12Claude Opus 5Anthropic93.72026-10-04
13Claude Opus 4.8Anthropic93.62026-10-04
14GPT-5.5OpenAI93.62026-08-23
15Kimi K3Moonshot93.52026-08-23
16Qwen3.8 2.4T A95BAlibaba93.52026-08-14
17GPT-5.2 ProOpenAI93.22026-08-23
18Grok4.5SpaceXAI93.12026-08-23
19MiniMax M3MiniMax932026-07-17
20GPT-5.4OpenAI932026-10-08
21GPT-5.6 TerraOpenAI92.92026-08-23
22Qwen3.8 Max PreviewAlibaba92.82026-09-16
23Gemini 3.6 FlashGoogle92.82026-09-10
24Claude Fable 5Anthropic92.62026-10-04
25Agnes 3.0 FlashSapiens AI92.42026-09-12
26GPT-5.2OpenAI92.42026-08-24
27DeepSeek-V4-ProDeepSeek92.42026-10-01
28Qwen3.7 MaxAlibaba92.42026-08-23
29Qwen3.8 Flash NextAlibaba92.32026-08-28
30Hy4Tencent92.32026-08-31
31GPT-5.6 LunaOpenAI92.32026-08-23
32Gemini 3.5 FlashGoogle92.22026-09-10
33Gemini 3 ProGoogle91.92026-08-24
34GLM-5.3Z.ai91.72026-09-10
35GPT-5.3 CodexOpenAI91.52026-05-02
36Claude Opus 4.6Anthropic91.32026-10-08
37GLM 5.3 FlashZ.ai91.22026-08-27
38GLM-5.2Z.ai91.22026-08-23
39Claude Sonnet 5Anthropic91.12026-10-04
40Kimi K2.6Moonshot91.12026-08-24
41DeepSeek-V4.1-FlashDeepSeek90.92026-09-11
42DeepSeek-V4-FlashDeepSeek90.82026-10-01
43Agnes 2.5 Pro BetaSapiens AI90.52026-08-27
44Qwen3.8 27BAlibaba90.52026-08-26
45Muse Spark 1.2Meta90.42026-08-06
46Gemini 3 Flash PreviewGoogle90.42026-08-23
47Qwen3.6 PlusAlibaba90.42026-10-08
48Hy3-previewTencent90.42026-08-23
49Qwen3.7 Plus PreviewAlibaba90.32026-08-26
50Grok 4.3SpaceXAI90.12026-05-23
51Claude Sonnet 4.6Anthropic89.92026-10-04
52GPT-5.2 CodexOpenAI89.92026-05-02
53Muse Spark 1.1Meta89.82026-07-11
54Kimi K2.7 CodeMoonshot89.62026-06-17
55Muse SparkMeta89.52026-08-23
56Inkling-SmallThinking Machines89.52026-08-23
57Qwen3.5 397B A17BAlibaba89.32026-08-24
58Nex-N2-ProNex AGI89.22026-06-26
59Solar Pro4Upstage89.12026-08-23
60Seed 2.0ByteDance88.92026-09-04
61Qwen3.6 Max PreviewAlibaba88.82026-05-02
62Grok 4.20SpaceXAI88.52026-09-10
63InklingThinking Machines88.32026-09-10
64GPT-5.1 InstantOpenAI88.12026-08-23
65GPT-5.1OpenAI88.12026-08-23
66GPT-5.4 miniOpenAI882026-08-23
67Nemotron 3 Ultra 550B A55BNVIDIA87.92026-09-01
68Kimi K2.5Moonshot87.92026-10-08
69Qwen3.6 27BAlibaba87.82026-08-23
70Grok 4SpaceXAI87.72026-08-23
71Agnes 2.5 Pro AlphaSapiens AI87.62026-07-25
72Grok 4 FastSpaceXAI87.52026-08-23
73Qwen3 Max (Reasoning)Alibaba87.42026-09-22
74MiniMax M2.7MiniMax87.42026-10-08
75GPT-5OpenAI87.32026-09-24
76K2 Horizon 375B A23BMBZUAI87.32026-09-04
77DeepSeek-V3.2-SpecialeDeepSeek87.12026-05-02
78Claude Opus 4.5Anthropic872026-08-24
79MiMo V2 ProXiaomi87.02026-05-02
80Gemini 3.1 Flash Lite PreviewGoogle86.92026-08-23
81Motif-3-BetaMotif Technologies86.92026-07-21
82GLM-5.1Z.ai86.82026-08-24
83Qwen3.5 122B A10BAlibaba86.62026-08-23
84MiMo V2.5 ProXiaomi86.62026-09-22
85Gemini 2.5 ProGoogle86.42026-09-03
86Apodex 1.1Apodex86.42026-08-31
87Ling-3.0-flash-VLInclusionAI86.22026-09-11
88GLM-5Z.ai862026-10-08
89Qwen3.6 35B A3BAlibaba862026-08-23
90GPT-5.1 CodexOpenAI86.02026-05-02
91GLM-4.7Z.ai85.92026-08-23
92Qwen3.5 27BAlibaba85.82026-08-23
93Gemma 4 31BGoogle85.72026-10-08
94Solar Open2 250BUpstage85.72026-08-13
95A.X-K2SK Telecom85.72026-08-13
96GPT-5.5 InstantOpenAI85.62026-08-23
97MiMo V2 OmniXiaomi85.52026-05-02
98Ling-3.0-flashInclusionAI85.52026-08-06
99Grok 4.1 FastSpaceXAI85.32026-06-23
100MiniMax M2.5MiniMax85.22026-08-24
◆ IV — INDEPENDENTLY VERIFIED · ◆ AA — AGGREGATOR · ◆ VA — VENDOR · ◇ SA — SOURCE-ATTRIBUTED

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

GPQA (unspecified)

VARIANT34 models · best score per model
#ModelVendorScoreVLast seen
01GPT-5.2OpenAI92.42026-05-01
02Kimi K2.6Moonshot912026-06-06
03Claude Sonnet 4.6Anthropic89.92026-05-01
04DeepSeek-V4-FlashDeepSeek88.52026-06-06
05DeepSeek-V4-ProDeepSeek87.82026-06-06
06Qwen3.5 397B A17BAlibaba87.12026-06-06
07MiniMax M2.7MiniMax86.62026-06-06
08GLM-5.1Z.ai86.12026-06-06
09Qwen3.6 35B A3BAlibaba862026-05-03
10GPT-5OpenAI85.72026-05-15
11Gemma 4 31BGoogle84.32026-06-06
12Qwen3.5 35B A3BAlibaba84.22026-06-06
13Claude Sonnet 4.5Anthropic83.42026-06-12
14Gemma 4 26B A4BGoogle82.32026-06-06
15DeepSeek-V3.2DeepSeek79.92026-05-15

Also tracked, no published surface rows yet: GPQA Diamond (CoT)

THE FAMILY BOARD · EVERY PRINT LABELED · EVERY SCORE SOURCE-ATTRIBUTED ON ITS OWN LIVE BOARDAll leaderboardsThe frontier boardModels by vendor