CostUse

AI model prices

Pick a provider, then a model. Rates, source, and check date are on the model page.

Hyperbolic

Open-model inference API plus GPU marketplace with aggressive per-token rates.

SambaNova

SambaNova Cloud inference with high-throughput open models and public per-token rates.

Qwen

Alibaba Cloud Model Studio first-party Qwen models with international token rates.

FriendliAI

Fast open-model Model APIs with cache-aware per-token pricing.

Meta

Muse Spark multimodal reasoning models on the Meta Model API, billed per token.

Microsoft

First-party Microsoft AI (MAI) reasoning and coding models on Microsoft Foundry.

InclusionAI

Ling open-weight hybrid reasoning models served through third-party serverless APIs.

iFLYTEK

Spark X models from iFLYTEK's open platform, served on domestic Chinese compute.

Spark X2.5

Spark X2.5

$0.236 in · $0.886 out per 1M tokens

Nex AGI

Nex-N2 open agentic models from the Shanghai Innovation Institute alliance, via API hosts.

OpenRouter

Unified multi-provider model router with public per-token list rates.