Updated 4 November 2026 · Artificial Analysis

AI model API prices compared

#ModelIntelligencePrice / 1M tokensInput / 1MOutput / 1MSpeedContext
1Gemini 4 Carbon Google60$8.00$4.00$2092 tok/s—
2Claude Opus 5.5 Anthropic59$8.00$4.00$2097 tok/s1M
3GPT-6 Astra OpenAI57$20$10$5051 tok/s1.1M
4GPT-6.1 Sol OpenAI55$4.00$2.00$1062 tok/s1.1M
5Claude Sonnet 5.5 Anthropic55$4.00$2.00$10135 tok/s1M
6GPT-6.1 Terra OpenAI51$2.00$1.00$5.00108 tok/s1.1M
7Gemini 4 Argon Google50$8.00$4.00$2081 tok/s1M
8Claude Fable 5.1 Anthropic50$20$10$5052 tok/s1M
9Kimi K3.1 Pro Kimi · open weights47$4.00$2.00$1028 tok/s1M
10GPT-6 Sol OpenAI44$4.00$2.00$10—1.1M
11Muse Wonder Meta44$8.00$4.00$20130 tok/s1M
12Muse Spark 1.3 Meta37$2.00$1.25$4.25—1M
13Claude Haiku 5.5 Anthropic37$0.20$0.10$0.50236 tok/s1M
14Grok 4.7 SpaceXAI36$3.00$2.00$6.0064 tok/s500K
15Qwen3.8 Max Alibaba34$3.00$2.00$6.0037 tok/s1M
16DeepSeek V4.1 Flash 1017 DeepSeek · open weights33$0.53$0.30$1.20157 tok/s1M
17GPT-5.6 Terra OpenAI32$4.50$2.00$12—1.1M
18GLM-5.3 Z AI · open weights32$2.15$1.40$4.4063 tok/s1M
19Grok 4.6 SpaceXAI31$3.00$2.00$6.00—500K
20Gemini 3.8 Flash Google31$1.50$0.75$3.75116 tok/s1M
21MiMo-V2.6-Pro Xiaomi · open weights30$0.54$0.43$0.8739 tok/s1.1M
22Ling 3.1 Flash 1019 InclusionAI30$0.45$0.30$0.90——
23MiniMax-M3.1-Flash MiniMax · open weights29$0.26$0.15$0.60108 tok/s1M
24Step 5 Preview StepFun · open weights29$1.43$1.00$2.70134 tok/s1M
25GPT-6 Luna OpenAI28$0.20$0.10$0.50126 tok/s1.1M
26GLM-5.3-Flash Z AI · open weights28$0.24$0.15$0.5063 tok/s1M
27MiMo-V2.6-Flash Xiaomi · open weights27$0.17$0.14$0.2830 tok/s1.1M
28Mistral Large 4 Mistral · open weights26$1.05$0.70$2.10—1M
29Mistral Large 4 Preview Mistral25$1.05$0.70$2.10——
30Muse Spark Open Meta · open weights25$2.00$1.25$4.25192 tok/s1M
31GPT-5.6 Luna OpenAI24$0.45$0.20$1.20—1.1M
32DeepSeek V4 Pro 0813 DeepSeek · open weights23$1.98$1.32$3.9673 tok/s1M
33Qwen3.8 27B Alibaba · open weights20$1.13$0.50$3.0039 tok/s1M
34MiniMax-M3 MiniMax · open weights19$0.53$0.30$1.20102 tok/s1M
35Beam 1.0 Reflection18$0.60$0.40$1.20321 tok/s—
36Kimi K3 Kimi · open weights17$6.00$3.00$15—1M
37Nemotron 3.5 Super VL NVIDIA · open weights14$0.28$0.20$0.50428 tok/s1M
38Inkling Small Thinking Machines · open weights13$0.53$0.30$1.20201 tok/s524K
39Inkling Thinking Machines · open weights13$1.76$1.00$4.05196 tok/s524K
40Nemotron 3 Ultra 550B A55B NVIDIA · open weights12$1.13$0.60$2.70161 tok/s262K
41Gemini 3.5 Flash-Lite Google12$0.85$0.30$2.50374 tok/s1M
42Claude 4.5 Haiku Anthropic7$2.00$1.00$5.00——
43Muse Glimmer Meta6$0.53$0.30$1.20115 tok/s—
44Gemini 3.1 Flash-Lite Google6$0.56$0.25$1.50—1M
45Nemotron 3.5 Lightning NVIDIA · open weights6$0.10$0.06$0.20271 tok/s262K
46Granite 4.2 8B IBM · open weights5$0.11$0.06$0.2575 tok/s131K
47Mistral Medium 3.5 Mistral5$3.00$1.50$7.50169 tok/s262K
48Mistral Small 4 Mistral · open weights4$0.26$0.15$0.60163 tok/s262K

Blended price is 3 parts input to 1 part output, the cheapest hosted rate tracked by Artificial Analysis. Providers change prices often; check the provider before you build on a number.

Data: Artificial Analysis (intelligence and coding indexes, prices, speed) and OpenRouter (open weights, context windows). Updated 4 November 2026; refreshed every 6 hours.

More rankings