LLM API providers
Every provider we track, with how many models each publishes and what they cost. Prices are USD per 1M tokens on the standard tier, re-read daily.
OpenAI pricing
71 modelsOpenAI prices most models in short- and long-context tiers, and offers cached input at a large discount for repeated prompt prefixes.
Input from Free to $150.00 per 1M tokens
Qwen pricing
53 modelsby Alibaba
Alibaba ships the widest range of sizes of any provider here, from sub-cent flash models to flagship Max tiers.
Input from $0.030 to $4.00 per 1M tokens
Gemini pricing
31 modelsby Google
Google publishes a free tier alongside paid pricing, and several Gemini models charge a higher rate above a long-context threshold.
Input from $0.100 to $3.50 per 1M tokens
GLM pricing
21 modelsby Zhipu AI
Zhipu publishes several genuinely free Flash models alongside paid GLM tiers, which is unusual among hosted APIs.
Input from Free to $2.20 per 1M tokens
Claude pricing
18 modelsby Anthropic
Anthropic quotes prices per million tokens (MTok) and bills prompt caching as separate write and read rates, with cache hits far cheaper than base input.
Input from $0.800 to $15.00 per 1M tokens
DeepSeek pricing
18 modelsDeepSeek is among the cheapest capable APIs, with aggressive cache-hit pricing that rewards repeated prompt prefixes.
Input from $0.030 to $0.800 per 1M tokens
Grok pricing
8 modelsby xAI
xAI publishes a machine-readable model catalogue, including context window and a long-context tier that applies above 128K tokens.
Input from $1.00 to $2.00 per 1M tokens
Kimi pricing
8 modelsby Moonshot AI
Moonshot’s Kimi models target long-context work at a fraction of Western flagship pricing.
Input from $0.450 to $0.950 per 1M tokens
Doubao pricing
1 modelsby ByteDance
ByteDance sells Doubao through Volcengine Ark, priced well below Western equivalents.
Input from $0.100 to $0.100 per 1M tokens
ERNIE pricing
1 modelsby Baidu
Baidu sells ERNIE through the Qianfan platform.
Input from $0.420 to $0.420 per 1M tokens