CostUse

AI model prices

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

Together AI

Open-model serverless inference, image, audio, and dedicated GPU endpoints.

Perplexity

Search-grounded Sonar models and agent tools with token plus request fees.

Sonar

Sonar

$1.00 in · $1.00 out per 1M tokens

Sonar Pro

Sonar Pro

$3.00 in · $15.00 out per 1M tokens

Cerebras

Wafer-scale inference with extreme tokens-per-second on open models.

AI21

Jamba hybrid models for long-context enterprise chat and document work.

Voyage AI

Specialized embedding and rerank models priced per million tokens.

voyage-4

voyage-4

$0.06 in · $0 out per 1M tokens

DeepInfra

Low-cost serverless inference for open models with transparent per-token rates.

Moonshot AI

Kimi models with million-token context and first-party cache-aware pricing.

Kimi K3

Kimi K3

$3.00 in · $15.00 out per 1M tokens

Kimi K2.6

Kimi K2.6

$0.95 in · $4.00 out per 1M tokens

Kimi K2.5

Kimi K2.5

$0.6 in · $3.00 out per 1M tokens

Z.AI

GLM flagship models from Z.AI with first-party cache-aware token pricing.

GLM-5

GLM-5

$1.00 in · $3.20 out per 1M tokens

GLM-5.2

GLM-5.2

$1.40 in · $4.40 out per 1M tokens

GLM-5.3

GLM-5.3

$1.40 in · $4.40 out per 1M tokens

Novita AI

Open-model serverless API with competitive per-token rates across Llama, Qwen, and DeepSeek.

Nebius

European AI cloud Token Factory with transparent per-token open-model inference.