VECTOR WIREAI INTELLIGENCE
⌘K
NVDA$1,847+3.2%MSFT$512+1.1%GOOGL$199-0.4%META$728+2.7%AMD$184-1.2%TSM$212+0.6%PLTR$98+4.1%AI IDX4,821+1.9%
PKT
SEEDRefresh Models —Deals —Regulatory —Sources —

arc_agi_1

78 models tracked

ARC-AGI-1 — Chollet's abstraction corpus v1; distinct from ARC-AGI-2/3, unrelated to arc_challenge (AI2 ARC)

#ModelVendorBest scoreRunsLast seen
1Claude Fable 5 (Extra High)Anthropic98.512026-08-06
2Gemini 3.1 ProGoogle9832026-06-21
3Opus 5Anthropic97.532026-07-25
4GPT-5.6 SolOpenAI97.522026-07-24
5GPT-5.6 TerraOpenAI96.512026-07-10
6Gemini 3 Deep ThinkGoogle9622026-05-10
7GPT-5.5OpenAI9532026-05-10
8GPT-5.5 ProOpenAI9512026-05-10
9GPT-5.2 (Refine.)OpenAI94.512026-05-10
10GPT-5.4 ProOpenAI94.532026-06-21
11GPT-5.4OpenAI93.752026-06-21
12Claude Opus 4.6Anthropic9332026-06-21
13Claude Opus 4.8Anthropic92.522026-07-24
14Gemini 3.5 FlashGoogle92.522026-05-29
15Claude Opus 4.7Anthropic9252026-06-21
16GPT-5.2 ProOpenAI90.512026-05-10
17Claude Fable 5Anthropic90.512026-08-06
18Grok 4.20SpaceXAI89.512026-05-10
19GPT-5.6 LunaOpenAI8882026-08-12
20Claude Sonnet 4.6Anthropic86.532026-06-19
21GPT-5.2OpenAI86.242026-05-10
22Gemini 3.7 FlashGoogle85.212026-08-21
23Gemini 3 Flash PreviewGoogle84.752026-05-29
24Inkling-SmallThinking Machines8432026-07-31
25DeepSeek-V4-FlashDeepSeek8422026-08-09
26Claude Opus 4.5Anthropic8242026-07-29
27Grok 4 (Refine.)SpaceXAI79.612026-05-10
28InklingThinking Machines79.522026-07-30
29Grok4.5SpaceXAI79.212026-07-24
30GLM-5.2 Full Open SourceZ.ai7722026-07-13
31Gemini 3 ProGoogle7512026-05-10
32Grok 4.6SpaceXAI74.812026-08-14
33GPT-5.1OpenAI72.842026-06-20
34GPT-5 ProOpenAI70.212026-05-10
35Grok 4SpaceXAI66.712026-05-10
36Kimi K3Moonshot65.712026-08-01
37Kimi K2.5Moonshot65.312026-05-10
38Claude Sonnet 4Anthropic63.722026-05-10
39MiniMax M2.5MiniMax63.712026-05-10
40Claude Sonnet 4.5Anthropic63.722026-05-10
41o3OpenAI60.812026-05-10
42o3-proOpenAI59.312026-05-10
43o4-miniOpenAI58.732026-05-10
44GPT-5.4 miniOpenAI5822026-05-15
45DeepSeek-V3.2DeepSeek5712026-05-10
46GPT-5OpenAI56.242026-05-10
47GPT-5 miniOpenAI54.342026-05-15
48Grok 4 FastSpaceXAI48.512026-05-15
49Claude Haiku 4.5Anthropic47.712026-05-10
50GLM-5Z.ai44.712026-05-10
51Gemini 2.5 ProGoogle4122026-05-10
52Tiny Recursion Model (TRM)Bespoke4012026-05-10
53Gemini 3.6 FlashGoogle34.512026-08-07
54o3-miniOpenAI34.522026-05-10
55Gemini 2.5 FlashGoogle33.322026-05-10
56GPT-5.4 nanoOpenAI3322026-05-10
57Hierarchical Reasoning Model (HRM)Bespoke3212026-05-10
58Codex MiniOpenAI27.312026-05-10
59Claude 4 OpusAnthropic2722026-05-10
60DeepSeek-R1DeepSeek21.222026-05-10
61Claude 3.7Anthropic21.212026-07-14
62GPT-5 nanoOpenAI20.732026-05-10
63Grok 3 Mini ReasoningSpaceXAI16.512026-05-10
64OpenAI o1-miniOpenAI1412026-05-10
65Claude 3.7 SonnetAnthropic13.612026-05-10
66Qwen3 235B A22B Instruct 2507Alibaba1122026-06-19
67GPT-4.5 PreviewOpenAI10.312026-05-10
68Gemini 3.5 Flash LiteGoogle7.512026-08-07
69Magistral Medium 1Mistral6.122026-05-10
70MagistralMistral6.112026-08-03
71GPT-4.1OpenAI5.512026-05-10
72Grok 3SpaceXAI5.522026-06-19
73Magistral SmallMistral512026-05-10
74GPT-4oOpenAI4.512026-05-10
75Llama 4 MaverickMeta4.412026-05-10
76GPT-4.1 miniOpenAI3.522026-06-19
77Llama 4 ScoutMeta0.522026-06-19
78GPT-4.1 nanoOpenAI022026-06-19

Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.