By provider

AI model providers

One page per lab — its full lineup, cheapest pick, biggest context window and any models on the deprecation clock. Prices sourced from official docs.

OpenAI29 models →

Maker of GPT and the o-series reasoning models, available through the OpenAI API and Azure.

Cheapest GA · $0.463/MContext · 1.1M13 on deprecation watch
Anthropic16 models →

Maker of the Claude family — frontier models tuned for coding, agents and long-context reasoning.

Cheapest GA · $2/MContext · 1M7 on deprecation watch
Google9 models →

Maker of the Gemini family, served via the Gemini API (Google AI Studio) and Vertex AI.

Cheapest GA · $0.563/MContext · 1.0M6 on deprecation watch
xAI17 models →

Maker of the Grok family, served via the xAI API.

Cheapest GA · $1.25/MContext · 1M8 on deprecation watch
Mistral20 models →

European lab shipping both open-weight and commercial models via La Plateforme.

Cheapest GA · $0.04/MContext · 256K12 on deprecation watch
Alibaba18 models →

Maker of the Qwen family — open-weight and hosted via Alibaba Cloud Model Studio.

Cheapest GA · $0.138/MContext · 1M5 on deprecation watch
Amazon6 models →

Maker of the Nova family on Amazon Bedrock, plus a managed gateway to third-party models.

Cheapest GA · $0.061/MContext · 1M1 on deprecation watch
Moonshot17 models →

Maker of the Kimi family — long-context and agentic open-weight models.

Cheapest GA · $0.65/MContext · 262K7 on deprecation watch
Perplexity5 models →

Maker of the Sonar search-grounded models, served via the Perplexity API.

Cheapest GA · $1/MContext · 200K1 on deprecation watch
DeepSeek4 models →

Open-weight reasoning and chat models known for aggressive pricing.

Context · 1M2 on deprecation watch
Meta12 models →

Maker of the open-weight Llama family, run on your own infrastructure or hosted providers.

Cheapest GA · $0.022/MContext · 10M
Cohere28 models →

Enterprise-focused lab known for the Command models plus embeddings and rerank.

Cheapest GA · $0.066/MContext · 256K10 on deprecation watch