💬 General Chat

General capabilities — Intelligence Index, MMLU-Pro.

オープンソース100/100 件のモデル

#			アクセス
1	Claude Opus 5.5 Anthropic	99.7	試す
2	Claude Sonnet 5.5 Anthropic	97.4	試す
3	Claude Fable 5.1 Anthropic	92.8	試す
4	Gemini 4 Argon Google	92.1	試す
5	GPT-6 Astra OpenAI	91.7	試す
6	GPT-6.1 Sol OpenAI	90.9	試す
7	Claude Opus 5 Anthropic	88.0	試す
8	Claude Fable 5 Anthropic	85.2	試す
9	Muse Spark 1.3 Meta	84.6	試す
10	GPT-6 Sol OpenAI	84.2	試す
11	Grok 4.7 SpaceXAI	82.3	–
12	MiMo-V2.6-Pro Xiaomi	82.1	–
13	GPT-5.6 Sol OpenAI	81.7	試す
14	Qwen3.8 Max Alibaba	80.4	試す
15	GLM-5.3 Z AI	79.1	試す
16	Grok 4.6 SpaceXAI	78.2	–
17	Step 5 Preview StepFun	78.0	試す
18	Kimi K3 Kimi	76.6	試す
19	GLM 5.3 Flash Z AI	74.6	試す
20	Ling 3.1 Flash InclusionAI	74.2	–
21	GPT-5.6 Terra OpenAI	74.1	試す
22	Gemini 3.8 Flash Google	73.3	試す
23	Claude Opus 4.8 Anthropic	72.8	試す
24	Qwen3.8-Flash-Next Alibaba	71.5	試す
25	Qwen3.8 2.4T A95B Alibaba	71.4	試す
26	DeepSeek V4.1 Flash DeepSeek	71.3	試す
27	Gemini 3.7 Flash Google	70.9	試す
28	Muse Spark 1.2 Meta	70.8	試す
29	Claude Opus 4.7 Anthropic	70.3	試す
30	GPT-6 Luna OpenAI	69.4	試す
31	MiMo-V2.6-Flash Xiaomi	69.1	–
32	Grok 4.5 SpaceXAI	69.0	試す
33	Claude Sonnet 5 Anthropic	67.9	試す
34	GPT-5.5 OpenAI	66.8	試す
35	GPT-5.4 OpenAI	66.8	試す
36	GPT-5.6 Luna OpenAI	66.6	試す
37	DeepSeek V4 Pro 0813 DeepSeek	65.3	試す
38	DeepSeek V4 Flash Vision DeepSeek	63.6	試す
39	DeepSeek V4 Flash 0731 DeepSeek	62.4	試す
40	Qwen3.8 27B Alibaba	61.8	試す
41	Gemini 3.6 Flash Google	61.7	試す
42	Motif 3 Motif Technologies	61.6	–
43	JT-4.1 Flash 236B A21B China Mobile	61.3	–
44	Muse Spark 1.1 Meta	61.0	試す
45	GLM-5.2 Z AI	60.6	試す
46	Gemini 3.5 Flash Google	59.9	試す
47	K2 Horizon 375B A23B Institute of Foundation Models	57.2	–
48	GPT-5.3 Codex OpenAI	56.1	試す
49	Muse Spark Meta	55.5	試す
50	Claude Opus 4.6 Anthropic	55.2	試す
51	DeepSeek V4 Pro 0424 DeepSeek	54.4	試す
52	Qwen3.7 Max Alibaba	53.5	試す
53	MiniMax-M3 MiniMax	53.3	試す
54	Solar Pro 4 Upstage	53.0	–
55	Claude Sonnet 4.6 Anthropic	52.6	試す
56	Gemini 3.1 Pro Preview Google	52.0	試す
57	Nex-N2-Pro Nex AGI	51.7	–
58	GPT-5.2 OpenAI	51.7	試す
59	Qwen3.6 Max Preview Alibaba	51.2	試す
60	MiMo-V2-Pro Xiaomi	50.8	–
61	Quasar 438B Multiverse Computing	50.8	–
62	Apodex 1.1 Apodex	50.7	–
63	Grok Build 0.1 0616 SpaceXAI	50.5	–
64	Claude Opus 4.5 Anthropic	49.3	試す
65	K2 Horizon MoVA 36B A4B Institute of Foundation Models	49.1	–
66	GLM-5 Z AI	49.0	試す
67	Kimi K2.6 Kimi	49.0	試す
68	Inkling Small Thinking Machines	49.0	–
69	GPT-5.5 Instant OpenAI	48.8	試す
70	GPT-5.2 Codex OpenAI	48.7	試す
71	Qwen3.6 Plus Alibaba	48.6	試す
72	Kimi K2.7 Code Kimi	48.2	試す
73	Ling-3.0-flash-VL InclusionAI	48.1	–
74	Hy3 Tencent	47.9	–
75	Solar Open2 250B Upstage	47.7	–
76	GLM-5-Turbo Z AI	47.7	試す
77	Inkling Thinking Machines	47.6	–
78	Solar Mini 4 Upstage	47.6	–
79	Gemini 3 Pro Preview Google	47.5	試す
80	MiMo-V2.5-Pro Xiaomi	47.5	–
81	GLM-5.1 Z AI	47.3	試す
82	Qwen3.7 Plus Alibaba	47.0	試す
83	Grok 4.20 0309 v2 SpaceXAI	46.7	–
84	MiMo-V2.5 Xiaomi	46.2	–
85	Grok 4.3 SpaceXAI	45.9	–
86	MiMo-V2-Omni-0327 Xiaomi	45.6	–
87	Gemini 3 Flash Preview Google	45.4	試す
88	Grok 4.20 0309 SpaceXAI	45.4	試す
89	DeepSeek V4 Flash 0420 DeepSeek	45.0	試す
90	Ling-3.0-flash-Fin InclusionAI	45.0	–
91	A.X-K2 SK Telecom	44.6	–
92	GPT-5.4 mini OpenAI	43.8	試す
93	MiMo-V2-Omni Xiaomi	43.5	–
94	Nemotron 3 Ultra 550B A55B NVIDIA	43.5	試す
95	Gemini 3.5 Flash-Lite Google	43.4	試す
96	G9v3-39A5B AI9Stars	43.4	–
97	GLM 5V Turbo Z AI	43.2	試す
98	Hy3-preview Tencent	42.4	–
99	GPT-5.1 OpenAI	42.3	試す
100	Kimi K2.5 Kimi	41.9	試す

スコア算出方法 — AIM Index

● 能力 85%● 最新性 15%

LMArenaArtificial Analysis

データソース (1)

Artificial Analysis ↗