模型排行

评分方法

来源为 Artificial Analysis 当前公开榜单。综合能力仅展示非估算的 AA 智能指数，速度单位为输出 tokens/s；不同推理配置分别列示。单价来自 OpenRouter 的同名基础模型，完整任务成本可能随推理量变化。

本次展示 134 个模型配置 · AA 公开当前模型榜单

左右滑动表格，查看评分、价格与操作 →

排名	模型 / 厂商	智能指数Artificial Analysis	上下文tokens	输入 / 输出USD / 百万 tokens	评测来源
26	Grok 4.7SpaceXAI high	46	500K	— / —	Artificial Analysis
27	MiMo-V2.6-ProXiaomi	46	1M	$0.44 / $0.87	Artificial Analysis
28	GPT-6 AstraOpenAI low	46	1M	$10 / $50含阶梯 / 时段定价	Artificial Analysis
29	Qwen3.8 MaxAlibaba 0902	45	984K	$2 / $6	Artificial Analysis
30	Muse Spark 1.3Meta xhigh	45	1M	— / —	Artificial Analysis
31	GLM-5.3Z AI max	45	1M	— / —	Artificial Analysis
32	Step 5 PreviewStepFun	44	1M	$1 / $2.7	Artificial Analysis
33	Kimi K3Kimi max	44	1.05M	— / —	Artificial Analysis
34	Claude Haiku 5.5Anthropic max	43	1M	$0.1 / $0.5含阶梯 / 时段定价	Artificial Analysis
35	Claude Opus 5.5Anthropic low with fallback	42	1M	$4 / $20	Artificial Analysis
36	Grok 4.7SpaceXAI low	42	500K	— / —	Artificial Analysis
37	GPT-6.1 SolOpenAI low	42	1M	$2 / $10含阶梯 / 时段定价	Artificial Analysis
38	GPT-5.6 TerraOpenAI max	42	1M	$2 / $12含阶梯 / 时段定价	Artificial Analysis
39	GLM-5.3-FlashZ AI	42	1M	— / —	Artificial Analysis
40	Claude Haiku 5.5Anthropic xhigh	41	1M	$0.1 / $0.5含阶梯 / 时段定价	Artificial Analysis
41	Ling 3.1 FlashInclusionAI	41	1M	$0.08 / $0.23	Artificial Analysis
42	Gemini 3.8 FlashGoogle high	41	1M	$0.83 / $4.13	Artificial Analysis
43	Claude Sonnet 5.5Anthropic medium with fallback	41	1M	$2 / $10	Artificial Analysis
44	Qwen3.8 2.4T A95BAlibaba	40	262K	$2 / $6	Artificial Analysis
45	Qwen3.8-Flash-NextAlibaba	40	256K	— / —	Artificial Analysis
46	Gemini 3.8 FlashGoogle medium	40	1M	$0.83 / $4.13	Artificial Analysis
47	DeepSeek V4.1 FlashDeepSeek max	39	1M	$0.04 / $0.08	Artificial Analysis
48	Muse Spark OpenMeta xhigh	39	1.05M	— / —	Artificial Analysis
49	Mistral Large 4 PreviewMistral	38	524K	— / —	Artificial Analysis
50	GPT-6 LunaOpenAI max	38	1M	$0.1 / $0.5含阶梯 / 时段定价	Artificial Analysis

没有可靠匹配的价格或规格时显示「—」，保留数据缺口。只比较 API 价格