AI model API pricing

33 models

One endpoint and one key for every model currently available on AnyModel. The bold price is the live rate you actually pay per 1M tokens; input and output are shown separately when they differ. The vendor's list price is shown next to it for comparison.

ModelProviderAnyModel /1MVendor list /1MContext
GPT-5.6 SolPopularcx/gpt-5.6-solOpenAI$0.1$5 / $301M
GPT-5.6 TerraPopularcx/gpt-5.6-terraOpenAI$0.05$2.5 / $151M
GPT-5.6 Lunacx/gpt-5.6-lunaOpenAI$0.025$1 / $6400K
GPT-5.5Popularcx/gpt-5.5OpenAI$0.05$5 / $301M
GPT-5.4Popularcx/gpt-5.4OpenAI$0.038$2.5 / $151M
GPT-5.4 minicx/gpt-5.4-miniOpenAI$0.025$0.75 / $4.5400K
Claude Opus 4.8Popularcc/claude-opus-4-8Anthropic$0.4$5 / $251M
Claude Sonnet 4.6Popularag/claude-sonnet-4-6Anthropic$0.24$3 / $151M
Claude Sonnet 5cc/claude-sonnet-5Anthropic$0.16$2 / $101M
Claude Opus 4.7cc/claude-opus-4-7Anthropic$0.4$5 / $251M
Claude Opus 4.6cc/claude-opus-4-6Anthropic$0.4$5 / $251M
Claude Haiku 4.5cc/claude-haiku-4-5-20251001Anthropic$0.08$1 / $5200K
Gemini 3.5 Flashag/gemini-3-flash-agentGoogle$0.03$1.5 / $91M
Gemini 3.1 Flash-Litegc/gemini-3.1-flash-lite-previewGoogle$0.03$0.25 / $1.51M
Free Models Auto Routeram/freeOTHER$0 / $0Varies
Qwen 3.6 35B A3Bam/qwen3.6-35b-a3bAlibaba$0.16 / $1.151M
MiniMax M3am/minimax-m3MiniMax$0.35 / $1.381M
DeepSeek V4 Proam/deepseek-v4-proDeepSeek$0.5 / $11M
DeepSeek V4 Flasham/deepseek-v4-flashDeepSeek$0.09 / $0.181M
GLM-5.2am/glm-5.2Zhipu$1.07 / $3.45$1.4 / $4.41M
GLM-5.2glm/glm-5.2Zhipu$0.1$1.4 / $4.41M
Kimi K3kmc/k3Moonshot$0.075$3 / $15256K
Kimi K2.7 Codekmc/kimi-for-codingMoonshot$0.038256K
Gemini 3 Flashag/gemini-3-flashGoogle$0.03
Gemini 3.1 Proag/gemini-pro-agentGoogle$0.085
GPT-OSS 120Bag/gpt-oss-120b-mediumOpenAI$0.005
Gemini 2.5 Flashgc/gemini-2.5-flashGoogle$0.03
Gemini 2.5 Flash Litegc/gemini-2.5-flash-liteGoogle$0.03
Gemini 2.5 Progc/gemini-2.5-proGoogle$0.085
GLM 4.6Vglm/glm-4.6vZhipu$0.015
GLM 4.7glm/glm-4.7Zhipu$0.015
GLM 5glm/glm-5Zhipu$0.025
GLM-5.1glm/glm-5.1Zhipu$0.025

AnyModel prices are shown per 1M tokens. Most models use one flat rate; models with separate input/output billing show both rates. Vendor list prices are shown only when directly comparable. No subscription, no minimums.

One key, every model, pay per token

Point any OpenAI-compatible client at AnyModel and switch models by name — no separate billing per provider. Start free with trial tokens, then top up and pay only for what you use.