All models
All models
18 active models across our network. Click any for full pricing, capabilities, and code samples.
Language models (text)
Anthropic
Claude Opus 4.7
Anthropic's most capable model.
$18.00/1M in1000K ctx
Anthropic
Claude Sonnet 4.6
Balanced speed + capability.
$3.60/1M in200K ctx
OpenAI
GPT-5
OpenAI's flagship general-purpose model.
$3.00/1M in256K ctx
OpenAI
GPT-5 mini
Fast, cheap, still smart.
$0.18/1M in128K ctx
Groq
Llama 3.3 70B (Groq)
Llama at Groq speed.
$0.71/1M in128K ctx
Anthropic
Claude Haiku 4.5
Fast and cheap from Anthropic.
$0.96/1M in200K ctx
Mistral
Codestral
Low-latency code completion from Mistral.
$0.24/1M in32K ctx
Cohere
Command R+
Cohere's flagship RAG-optimized model.
$3.00/1M in128K ctx
DeepSeek
DeepSeek R1
Reasoning model — rivals o1/o3 at 20x lower cost
$0.66/1M in66K ctxBeta
DeepSeek
DeepSeek V3
Open-weights reasoning champion, rock-bottom pricing.
$0.32/1M in128K ctxBeta
Google Gemini
Gemini 2.5 Pro
Google's top multimodal model.
$1.50/1M in2000K ctx
xAI
Grok 4
xAI's flagship model with real-time X data.
$3.60/1M in256K ctx
Mistral
Mistral Large
Mistral's flagship multilingual model.
$2.40/1M in128K ctx
Perplexity
Sonar Pro
Online LLM with live citations.
$3.60/1M in200K ctx