透明度与可获得性

开放性指数

从权重、许可、技术披露和复现条件观察模型开放程度。

数据快照：2026-07-21查看原始来源 ↗

主要指标Openness components

快照日期2026-07-21

公开评测完整榜单
排名	机构	模型	开放性指数	智能指数	模型可用性	模型透明度	预训练数据访问	预训练数据许可	后训练数据访问	后训练数据许可
1	Allen Institute for AI	Olmo 3.1 32B Think	88.89	8.09	6	10	3	1	3	1
2	Allen Institute for AI	Molmo 7B-D	88.89	3.81	6	10	3	1	3	1
3	Allen Institute for AI	Olmo 3 7B Instruct	88.89	2.81	6	10	3	1	3	1
4	Allen Institute for AI	Olmo 3.1 32B Instruct	88.89	6.47	6	10	3	1	3	1
5	Allen Institute for AI	Olmo 3 7B Think	88.89	3.97	6	10	3	1	3	1
6	MBZUAI Institute of Foundation Models	K2 Think V2	88.89	17.26	6	10	3	1	3	1
7	MBZUAI Institute of Foundation Models	K2-V2 (low)	88.89	8.55	6	10	3	1	3	1
8	MBZUAI Institute of Foundation Models	K2-V2 (high)	88.89	14.19	6	10	3	1	3	1
9	MBZUAI Institute of Foundation Models	K2-V2 (medium)	88.89	12.43	6	10	3	1	3	1
10	Swiss AI Initiative	Apertus 8B Instruct	88.89	1	6	10	3	1	3	1
11	Swiss AI Initiative	Apertus 70B Instruct	88.89	2.4	6	10	3	1	3	1
12	Allen Institute for AI	OLMo 2 7B	88.89	3.85	6	10	3	1	3	1
13	Allen Institute for AI	Olmo 3 32B Think	88.89	6.4	6	10	3	1	3	1
14	Allen Institute for AI	OLMo 2 32B	88.89	5.01	6	10	3	1	3	1
15	NVIDIA	NVIDIA Nemotron 3 Super 120B A12B (Reasoning)	83.33	25.41	6	9	2	1	2	1
16	NVIDIA	Nemotron 3 Ultra 550B A55B (Reasoning)	83.33	37.76	6	9	2	1	2	1
17	NVIDIA	NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)	83.33	7.39	6	9	2	1	2	1
18	NVIDIA	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)	83.33	14.18	6	9	2	1	2	1
19	OpenBMB	MiniCPM5-1B (Non-reasoning)	83.33	11.74	6	9	3	3	3	3
20	OpenBMB	MiniCPM5-1B (Reasoning)	83.33	11.96	6	9	3	3	3	3
21	NVIDIA	NVIDIA Nemotron Nano 9B V2 (Reasoning)	72.22	8.84	6	7	2	1	2	1
22	NVIDIA	NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)	72.22	4.58	6	7	2	1	2	1
23	NVIDIA	NVIDIA Nemotron Nano 9B V2 (Non-reasoning)	72.22	7.38	6	7	2	1	2	1
24	NVIDIA	NVIDIA Nemotron Nano 12B v2 VL (Reasoning)	72.22	8.96	6	7	2	1	2	1
25	Allen Institute for AI	Molmo2-8B	72.22	2.03	6	7	3	1	3	1
26	Kimi	Kimi Linear 48B A3B Instruct	61.11	8.53	6	5	1	0	1	0
27	IBM	Granite 4.1 8B	61.11	6.67	6	5	2	1	2	1
28	IBM	Granite 4.1 3B	61.11	4.7	6	5	2	1	2	1
29	IBM	Granite 4.1 30B	61.11	8.86	6	5	2	1	2	1
30	IBM	Granite 4.0 H 350M	55.56	1	6	4	2	1	2	1
31	IBM	Granite 4.0 H 1B	55.56	2.66	6	4	2	1	2	1
32	IBM	Granite 4.0 350M	55.56	1	6	4	2	1	2	1
33	IBM	Granite 4.0 H Small	55.56	5.24	6	4	2	1	2	1
34	IBM	Granite 4.0 Micro	55.56	2.37	6	4	2	1	2	1
35	IBM	Granite 4.0 1B	55.56	2.07	6	4	2	1	2	1
36	Baidu	ERNIE 4.5 300B A47B	55.56	9.02	6	4	0	0	0	0
37	Z AI	GLM-4.5-Air	55.56	16.52	6	4	0	0	0	0
38	Z AI	GLM-4.5 (Reasoning)	55.56	19.49	6	4	0	0	0	0
39	NVIDIA	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)	52.78	9.08	4	5.5	1	0	1	1
40	NVIDIA	Llama Nemotron Super 49B v1.5 (Reasoning)	52.78	12.42	4	5.5	1	0	1	1
41	NVIDIA	Llama Nemotron Super 49B v1.5 (Non-reasoning)	52.78	8.68	4	5.5	1	0	1	1
42	Xiaomi	MiMo-V2-Flash (Non-reasoning)	52.78	24.67	6	3.5	0	0	1	0
43	NVIDIA	Llama 3.3 Nemotron Super 49B v1 (Reasoning)	52.78	12.25	4	5.5	1	0	1	1
44	NVIDIA	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)	52.78	8.54	4	5.5	1	0	1	1
45	NVIDIA	Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)	52.78	8.47	4	5.5	1	0	1	1
46	Xiaomi	MiMo-V2-Flash (Reasoning)	52.78	31.2	6	3.5	0	0	1	0
47	Z AI	GLM-4.5V (Reasoning)	52.78	9.15	6	3.5	1	0	0	0
48	Z AI	GLM-4.5V (Non-reasoning)	52.78	7	6	3.5	1	0	0	0
49	Mistral	Magistral Small 1.2	50	11.29	6	3	0	0	1	1
50	DeepSeek	DeepSeek V4 Flash (Reasoning, High Effort)	50	37.36	6	3	0	0	0	0
51	DeepSeek	DeepSeek V4 Pro (Reasoning, High Effort)	50	43.11	6	3	0	0	0	0
52	DeepSeek	DeepSeek V4 Flash (Non-reasoning)	50	28.65	6	3	0	0	0	0
53	DeepSeek	DeepSeek V4 Flash (Reasoning, Max Effort)	50	40.28	6	3	0	0	0	0
54	DeepSeek	DeepSeek V4 Pro (Non-reasoning)	50	31.22	6	3	0	0	0	0
55	DeepSeek	DeepSeek V4 Pro (Reasoning, Max Effort)	50	44.27	6	3	0	0	0	0
56	Microsoft	Phi-4	50	4.87	6	3	1	0	1	0
57	Microsoft	Phi-4 Mini Instruct	50	6.01	6	3	1	0	1	0
58	Microsoft	Phi-4 Multimodal Instruct	50	4.53	6	3	1	0	1	0
59	Google	Gemma 3 12B Instruct	50	5.51	6	3	0	0	0	0
60	Google	Gemma 3 1B Instruct	50	1	6	3	0	0	0	0
61	Google	Gemma 3 4B Instruct	50	1.12	6	3	0	0	0	0
62	Google	Gemma 3n E2B Instruct	50	1	6	3	0	0	0	0
63	Google	Gemma 3 27B Instruct	50	7.42	6	3	0	0	0	0
64	Google	Gemma 3n E4B Instruct	50	1.19	6	3	0	0	0	0
65	DeepSeek	DeepSeek R1 0528 (May '25)	50	20.08	6	3	0	0	0	0
66	StepFun	Step 3.5 Flash	50	25.5	6	3	0	0	0	0
67	Z AI	GLM-5 (Reasoning)	50	39.5	6	3	0	0	0	0
68	Z AI	GLM-5 (Non-reasoning)	50	32.41	6	3	0	0	0	0
69	Alibaba	Qwen3 VL 30B A3B (Reasoning)	50	13.34	6	3	1	0	1	0
70	Alibaba	Qwen3 VL 235B A22B (Reasoning)	50	20.6	6	3	1	0	1	0
71	Alibaba	Qwen3 VL 235B A22B Instruct	50	14.31	6	3	1	0	1	0
72	Alibaba	Qwen3 VL 32B Instruct	50	11.06	6	3	1	0	1	0
73	Alibaba	Qwen3 VL 30B A3B Instruct	50	10.02	6	3	1	0	1	0
74	Alibaba	Qwen3 VL 8B (Reasoning)	50	10.58	6	3	1	0	1	0
75	Alibaba	Qwen3 VL 32B (Reasoning)	50	17.93	6	3	1	0	1	0
76	Alibaba	Qwen3 VL 4B Instruct	50	4.09	6	3	1	0	1	0
77	Alibaba	Qwen3 VL 4B (Reasoning)	50	7.9	6	3	1	0	1	0
78	Alibaba	Qwen3 VL 8B Instruct	50	8.42	6	3	1	0	1	0
79	Nous Research	Hermes 4 - Llama-3.1 70B (Non-reasoning)	47.22	6.9	4	4.5	1	0	2	0
80	Nous Research	Hermes 4 - Llama-3.1 405B (Non-reasoning)	47.22	8.8	4	4.5	1	0	2	0
81	Nous Research	Hermes 4 - Llama-3.1 405B (Reasoning)	47.22	9	4	4.5	1	0	2	0
82	Nous Research	Hermes 4 - Llama-3.1 70B (Reasoning)	47.22	9.97	4	4.5	1	0	2	0
83	DeepSeek	DeepSeek R1 0528 Qwen3 8B	47.22	10.37	6	2.5	0	0	1	0
84	ServiceNow	Apriel-v1.5-15B-Thinker	47.22	21.23	6	2.5	0	0	1	0
85	Google	Gemma 3 270M	44.44	2.41	6	2	0	0	0	0
86	TII UAE	Falcon-H1R-7B	44.44	9.79	4	4	1	0	1	0
87	NVIDIA	Llama 3.1 Nemotron Instruct 70B	44.44	7.64	4	4	0	0	1	1
88	LongCat	LongCat Flash Lite	44.44	17.22	6	2	0	0	0	0
89	OpenBMB	MiniCPM-V 4.6 1.3B	44.44	4.19	6	2	0	0	0	0
90	Arcee AI	Trinity Large Thinking	44.44	24.47	6	2	0	0	0	0
91	Z AI	GLM-5.2 (max)	44.44	51.09	6	2	0	0	0	0
92	Alibaba	Qwen3 Omni 30B A3B Instruct	44.44	5.11	6	2	0	0	0	0
93	Alibaba	Qwen3 Next 80B A3B (Reasoning)	44.44	16.72	6	2	0	0	0	0
94	Alibaba	Qwen3 Next 80B A3B Instruct	44.44	13.72	6	2	0	0	0	0
95	Alibaba	Qwen3 Omni 30B A3B (Reasoning)	44.44	9.63	6	2	0	0	0	0
96	InclusionAI	Ling-mini-2.0	44.44	3.75	6	2	0	0	0	0
97	Mistral	Devstral Small (Jul '25)	44.44	9.26	6	2	0	0	0	0
98	DeepSeek	DeepSeek V3.2 Exp (Non-reasoning)	44.44	21.33	6	2	0	0	0	0
99	DeepSeek	DeepSeek V3.2 Exp (Reasoning)	44.44	25.45	6	2	0	0	0	0
100	Kimi	Kimi K2	44.44	19.4	4	4	1	0	1	0
101	Z AI	GLM-5.1 (Reasoning)	44.44	40.16	6	2	0	0	0	0
102	Z AI	GLM-4.7 (Non-reasoning)	44.44	26.56	6	2	0	0	0	0
103	Z AI	GLM-5.1 (Non-reasoning)	44.44	35.37	6	2	0	0	0	0
104	Z AI	GLM-4.7-Flash (Non-reasoning)	44.44	15.52	6	2	0	0	0	0
105	Z AI	GLM-4.7-Flash (Reasoning)	44.44	22.89	6	2	0	0	0	0
106	Z AI	GLM-4.7 (Reasoning)	44.44	33.7	6	2	0	0	0	0
107	Z AI	GLM-4.6 (Non-reasoning)	44.44	22.98	6	2	0	0	0	0
108	Z AI	GLM-4.6 (Reasoning)	44.44	28.7	6	2	0	0	0	0
109	Alibaba	Qwen3 Coder 480B A35B Instruct	44.44	17.98	6	2	0	0	0	0
110	Alibaba	Qwen3 235B A22B 2507 (Reasoning)	44.44	19.61	6	2	0	0	0	0
111	Alibaba	Qwen3 235B A22B 2507 Instruct	44.44	18.16	6	2	0	0	0	0
112	Alibaba	Qwen3 Coder 30B A3B Instruct	44.44	13.61	6	2	0	0	0	0
113	Alibaba	Qwen3 30B A3B 2507 Instruct	44.44	9.06	6	2	0	0	0	0
114	Alibaba	Qwen3 4B 2507 Instruct	44.44	7.12	6	2	0	0	0	0
115	Alibaba	Qwen3 30B A3B 2507 (Reasoning)	44.44	14.44	6	2	0	0	0	0
116	Alibaba	Qwen3 4B 2507 (Reasoning)	44.44	11.96	6	2	0	0	0	0
117	InclusionAI	Ling-flash-2.0	44.44	9.74	6	2	0	0	0	0
118	InclusionAI	Ling-1T	44.44	12.75	6	2	0	0	0	0
119	ByteDance Seed	Seed-OSS-36B-Instruct	44.44	18.34	6	2	0	0	0	0
120	Cohere	North Mini Code	41.67	19.8	6	1.5	0	0	1	0
121	Alibaba	Qwen3 Coder Next	41.67	21.11	6	1.5	0	0	1	0
122	OpenAI	gpt-oss-20b (high)	38.89	14.89	6	1	0	0	0	0
123	OpenAI	gpt-oss-120b (high)	38.89	23.83	6	1	0	0	0	0
124	Meta	Llama 3.3 Instruct 70B	38.89	9.4	4	3	1	0	1	0
125	Meta	Llama 3.1 Instruct 405B	38.89	8.5	4	3	1	0	1	0
126	Meta	Llama 3.2 Instruct 90B (Vision)	38.89	6.23	4	3	1	0	1	0
127	Meta	Llama 3.2 Instruct 11B (Vision)	38.89	3.33	4	3	1	0	1	0
128	Google	Gemma 4 12B (Non-reasoning)	38.89	13.19	6	1	0	0	0	0
129	Google	Gemma 4 31B (Reasoning)	38.89	29.35	6	1	0	0	0	0
130	Google	Gemma 4 26B A4B (Non-reasoning)	38.89	20.1	6	1	0	0	0	0
131	Google	Gemma 4 31B (Non-reasoning)	38.89	21.77	6	1	0	0	0	0
132	Google	Gemma 4 E4B (Reasoning)	38.89	12.5	6	1	0	0	0	0
133	Google	Gemma 4 26B A4B (Reasoning)	38.89	25.69	6	1	0	0	0	0
134	Google	Gemma 4 E4B (Non-reasoning)	38.89	8.91	6	1	0	0	0	0
135	Google	Gemma 4 12B (Reasoning)	38.89	22	6	1	0	0	0	0
136	Google	Gemma 4 E2B (Non-reasoning)	38.89	6.41	6	1	0	0	0	0
137	Google	Gemma 4 E2B (Reasoning)	38.89	9.26	6	1	0	0	0	0
138	Mistral	Mistral Large 3	38.89	15.93	6	1	0	0	0	0
139	Mistral	Mistral Small 4 (Reasoning)	38.89	19.56	6	1	0	0	0	0
140	Mistral	Mistral Small 4 (Non-reasoning)	38.89	12.37	6	1	0	0	0	0
141	Perplexity	R1 1776	38.89	6.31	6	1	0	0	0	0
142	StepFun	Step 3.7 Flash	38.89	30.27	6	1	0	0	0	0
143	Reka AI	Reka Flash 3	38.89	4.06	6	1	0	0	0	0
144	Nous Research	DeepHermes 3 - Mistral 24B Preview (Non-reasoning)	38.89	5.31	6	1	0	0	0	0
145	Xiaomi	MiMo-V2.5-Pro	38.89	42.24	6	1	0	0	0	0
146	Xiaomi	MiMo-V2.5	38.89	37.2	6	1	0	0	0	0
147	Xiaomi	MiMo-V2.5-Pro (Non-reasoning)	38.89	27.86	6	1	0	0	0	0
148	Sarvam	Sarvam 30B (high)	38.89	6.64	6	1	0	0	0	0
149	Sarvam	Sarvam 105B (high)	38.89	11.94	6	1	0	0	0	0
150	Multiverse Computing	HyperNova 60B 2605	38.89	17.8	6	1	0	0	0	0
151	Deep Cogito	Cogito v2.1 (Reasoning)	38.89	—	6	1	0	0	0	0
152	Nex AGI	Nex-N2-Pro	38.89	40.96	6	1	0	0	0	0
153	Cohere	Command A+	38.89	22.51	6	1	0	0	0	0
154	AI21 Labs	Jamba Reasoning 3B	38.89	4.13	6	1	0	0	0	0
155	Alibaba	Qwen3.5 397B A17B (Reasoning)	38.89	33.68	6	1	0	0	0	0
156	Alibaba	Qwen3.5 397B A17B (Non-reasoning)	38.89	31.98	6	1	0	0	0	0
157	Alibaba	Qwen3.6 35B A3B (Non-reasoning)	38.89	24.16	6	1	0	0	0	0
158	Alibaba	Qwen3.5 122B A10B (Reasoning)	38.89	32.28	6	1	0	0	0	0
159	Alibaba	Qwen3.6 27B (Non-reasoning)	38.89	30.46	6	1	0	0	0	0
160	Alibaba	Qwen3.5 0.8B (Reasoning)	38.89	5.48	6	1	0	0	0	0
161	Alibaba	Qwen3.5 9B (Reasoning)	38.89	21.44	6	1	0	0	0	0
162	Alibaba	Qwen3.5 9B (Non-reasoning)	38.89	20.32	6	1	0	0	0	0
163	Alibaba	Qwen3.6 35B A3B (Reasoning)	38.89	31.65	6	1	0	0	0	0
164	Alibaba	Qwen3.6 27B (Reasoning)	38.89	37.05	6	1	0	0	0	0
165	Alibaba	Qwen3.5 35B A3B (Non-reasoning)	38.89	23.99	6	1	0	0	0	0
166	Alibaba	Qwen3.5 4B (Reasoning)	38.89	20.09	6	1	0	0	0	0
167	Alibaba	Qwen3.5 122B A10B (Non-reasoning)	38.89	27.6	6	1	0	0	0	0
168	Alibaba	Qwen3.5 2B (Non-reasoning)	38.89	5.62	6	1	0	0	0	0
169	Alibaba	Qwen3.5 2B (Reasoning)	38.89	7.64	6	1	0	0	0	0
170	Alibaba	Qwen3.5 0.8B (Non-reasoning)	38.89	3.31	6	1	0	0	0	0
171	Alibaba	Qwen3.5 4B (Non-reasoning)	38.89	16	6	1	0	0	0	0
172	InclusionAI	Ling-2.6-1T	38.89	26.05	6	1	0	0	0	0
173	InclusionAI	Ling 2.6 Flash	38.89	14.05	6	1	0	0	0	0
174	InclusionAI	Ring-flash-2.0	38.89	8.16	6	1	0	0	0	0
175	InclusionAI	Ring-2.6-1T	38.89	30.57	6	1	0	0	0	0
176	Mistral	Mistral Small 3.2	38.89	10.56	6	1	0	0	0	0
177	DeepSeek	DeepSeek V3.1 Terminus (Reasoning)	38.89	30.44	6	1	0	0	0	0
178	DeepSeek	DeepSeek V3.1 Terminus (Non-reasoning)	38.89	21.41	6	1	0	0	0	0
179	Alibaba	Qwen3.5 27B (Reasoning)	38.89	33.78	6	1	0	0	0	0
180	Alibaba	Qwen3.5 35B A3B (Reasoning)	38.89	29.26	6	1	0	0	0	0
181	Alibaba	Qwen3.5 27B (Non-reasoning)	38.89	29.31	6	1	0	0	0	0
182	InclusionAI	Ring-1T	38.89	16.16	6	1	0	0	0	0
183	DeepSeek	DeepSeek R1 Distill Llama 70B	36.11	9.93	4	2.5	0	0	1	0
184	Mistral	Mistral Medium 3.5	33.33	29.95	5	1	0	0	0	0
185	Liquid AI	LFM2 2.6B	33.33	2.71	4	2	0	0	0	0
186	Liquid AI	LFM2 8B A1B	33.33	1.78	4	2	0	0	0	0
187	Liquid AI	LFM2.5-8B-A1B	33.33	8.32	4	2	0	0	0	0
188	MiniMax	MiniMax-M3	33.33	44.44	4	2	0	0	0	0
189	Kimi	Kimi K2.6 (Non-reasoning)	33.33	34.58	4	2	0	0	0	0
190	Nous Research	DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)	33.33	2.29	5	1	0	0	0	0
191	Tencent	Hy3-preview (Reasoning)	33.33	33.58	5	1	0	0	0	0
192	Tencent	Hy3-preview (Non-reasoning)	33.33	26.1	5	1	0	0	0	0
193	Cohere	Command A	33.33	7.67	3	3	0	0	0	0
194	Liquid AI	LFM2 1.2B	33.33	1.15	4	2	0	0	0	0
195	Kimi	Kimi K2.5 (Reasoning)	33.33	35.37	4	2	0	0	0	0
196	Kimi	Kimi K2.6	33.33	44.22	4	2	0	0	0	0
197	Kimi	Kimi K2.5 (Non-reasoning)	33.33	29.4	4	2	0	0	0	0
198	Naver	HyperCLOVA X SEED Think (32B)	30.56	17.02	4	1.5	1	0	0	0
199	Meta	Llama 4 Maverick	27.78	14.27	4	1	0	0	0	0
200	Meta	Llama 4 Scout	27.78	10.04	4	1	0	0	0	0
201	Mistral	Magistral Medium 1.2	27.78	17.87	2	3	0	0	1	1
202	Liquid AI	LFM2.5-1.2B-Thinking	27.78	2.75	4	1	0	0	0	0
203	Liquid AI	LFM2.5-1.2B-Instruct	27.78	2.71	4	1	0	0	0	0
204	Liquid AI	LFM2.5-VL-1.6B	27.78	1.01	4	1	0	0	0	0
205	Liquid AI	LFM2 24B A2B	27.78	4.95	4	1	0	0	0	0
206	Kimi	Kimi K2.7 Code	27.78	41.95	4	1	0	0	0	0
207	LG AI Research	Exaone 4.0 1.2B (Reasoning)	27.78	2.9	3	2	0	0	0	0
208	LG AI Research	EXAONE 4.0 32B (Non-reasoning)	27.78	6.01	3	2	0	0	0	0
209	LG AI Research	K-EXAONE (Non-reasoning)	27.78	16.74	4	1	0	0	0	0
210	LG AI Research	EXAONE 4.5 33B	27.78	22.96	3	2	0	0	0	0
211	LG AI Research	Exaone 4.0 1.2B (Non-reasoning)	27.78	2.77	3	2	0	0	0	0
212	LG AI Research	EXAONE 4.5 33B (Non-reasoning)	27.78	—	3	2	0	0	0	0
213	LG AI Research	K-EXAONE (Reasoning)	27.78	24.7	4	1	0	0	0	0
214	LG AI Research	EXAONE 4.0 32B (Reasoning)	27.78	10.59	3	2	0	0	0	0
215	MiniMax	MiniMax-M2	27.78	28.31	4	1	0	0	0	0
216	MiniMax	MiniMax-M2.5	27.78	33.65	4	1	0	0	0	0
217	MiniMax	MiniMax-M2.1	27.78	31.36	4	1	0	0	0	0
218	Kimi	Kimi K2 0905	27.78	23.54	4	1	0	0	0	0
219	Kimi	Kimi K2 Thinking	27.78	32.7	4	1	0	0	0	0
220	AI21 Labs	Jamba 1.7 Mini	22.22	2.73	4	0	0	0	0	0
221	AI21 Labs	Jamba 1.7 Large	22.22	5.3	4	0	0	0	0	0
222	MiniMax	MiniMax-M2.7	22.22	38.13	3	1	0	0	0	0
223	Alibaba	Qwen3 Max	16.67	24.01	2	1	0	0	0	0
224	Alibaba	Qwen3 Max Thinking (Preview)	16.67	25.03	2	1	0	0	0	0
225	Anthropic	Claude 4.5 Haiku (Non-reasoning)	11.11	23.71	2	0	0	0	0	0
226	Anthropic	Claude 4.5 Haiku (Reasoning)	11.11	29.58	2	0	0	0	0	0
227	Amazon	Nova Micro	11.11	4.74	2	0	0	0	0	0
228	Amazon	Nova Premier	11.11	12.73	2	0	0	0	0	0
229	Upstage	Solar Pro 2 (Reasoning)	11.11	8.99	2	0	0	0	0	0
230	Upstage	Solar Pro 2 (Non-reasoning)	11.11	7.77	2	0	0	0	0	0
231	ByteDance Seed	Doubao Seed Code	11.11	25.98	2	0	0	0	0	0
232	OpenAI	GPT-5.1 (Non-reasoning)	11.11	20.4	2	0	0	0	0	0
233	OpenAI	GPT-5 (ChatGPT)	11.11	15.3	2	0	0	0	0	0
234	Google	Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)	11.11	13.1	2	0	0	0	0	0
235	Google	Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)	11.11	18.83	2	0	0	0	0	0
236	Anthropic	Claude Opus 4.5 (Non-reasoning)	11.11	34.71	2	0	0	0	0	0
237	Anthropic	Claude 4.5 Sonnet (Non-reasoning)	11.11	29.28	2	0	0	0	0	0
238	Anthropic	Claude 4.5 Sonnet (Reasoning)	11.11	36.42	2	0	0	0	0	0
239	Anthropic	Claude Opus 4.5 (Reasoning)	11.11	40.77	2	0	0	0	0	0
240	Mistral	Mistral Medium 3.1	11.11	14.68	2	0	0	0	0	0
241	Mistral	Devstral Medium	11.11	12.4	2	0	0	0	0	0
242	SpaceXAI	Grok 3 mini Reasoning (high)	11.11	22.5	2	0	0	0	0	0
243	SpaceXAI	Grok 4 Fast (Non-reasoning)	11.11	16.47	2	0	0	0	0	0
244	SpaceXAI	Grok 4.1 Fast (Non-reasoning)	11.11	16.88	2	0	0	0	0	0
245	Amazon	Nova Pro	11.11	7.67	2	0	0	0	0	0
246	Amazon	Nova Lite	11.11	6.92	2	0	0	0	0	0
247	OpenAI	o3	5.56	30.4	1	0	0	0	0	0
248	Google	Gemini 2.5 Pro	5.56	25.82	1	0	0	0	0	0
249	OpenAI	GPT-5 nano (minimal)	5.56	8	1	0	0	0	0	0
250	OpenAI	GPT-5 mini (high)	5.56	25.31	1	0	0	0	0	0
251	OpenAI	GPT-5 (high)	5.56	34.71	1	0	0	0	0	0
252	OpenAI	GPT-5.1 (high)	5.56	36.87	1	0	0	0	0	0
253	OpenAI	GPT-5 Codex (high)	5.56	36.12	1	0	0	0	0	0
254	OpenAI	GPT-5 (medium)	5.56	33.74	1	0	0	0	0	0
255	OpenAI	GPT-5 (low)	5.56	31.15	1	0	0	0	0	0
256	OpenAI	GPT-5 nano (medium)	5.56	19	1	0	0	0	0	0
257	OpenAI	GPT-5 mini (minimal)	5.56	14.25	1	0	0	0	0	0
258	OpenAI	GPT-5 mini (medium)	5.56	30.92	1	0	0	0	0	0
259	OpenAI	GPT-5 (minimal)	5.56	17.18	1	0	0	0	0	0
260	OpenAI	GPT-5 nano (high)	5.56	19.87	1	0	0	0	0	0
261	Google	Gemini 3 Pro Preview (high)	5.56	39.55	1	0	0	0	0	0
262	Google	Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)	5.56	15.13	1	0	0	0	0	0
263	Google	Gemini 2.5 Flash Preview (Sep '25) (Reasoning)	5.56	23.8	1	0	0	0	0	0
264	SpaceXAI	Grok 4	5.56	33.28	1	0	0	0	0	0
265	SpaceXAI	Grok 4 Fast (Reasoning)	5.56	27.37	1	0	0	0	0	0
266	SpaceXAI	Grok Code Fast 1	5.56	21.61	1	0	0	0	0	0
267	SpaceXAI	Grok 4.1 Fast (Reasoning)	5.56	30.62	1	0	0	0	0	0

这个评测关注什么

从权重、许可、技术披露和复现条件观察模型开放程度。

阅读时注意

评测结果受任务集合、评分方法和模型版本影响。本站保留原始口径，不用它生成新的综合排名。

适用场景

用于回答一个明确的问题：模型在这一类任务上的相对表现如何，而不是“哪个模型绝对最好”。