Offerings

Every active offering: a model as served by one host, with its current canonical price. Sort by clicking a column header; filter with the row below. Prices are USD per 1M tokens unless the unit toggle is set to per 1M characters (an approximation using a measured, blended tokens-per-character ratio).

Showing tier-tagged and vendor-direct models. Show all offerings, including the long tail

More columns: + Vendor + Region + Cache + Batch + Tokenizer + Last change - Source All columns Default columns

ModelHostVariantRegionInputOutputBlended ↑?ContextConfidence?Source
Mistral Large 2.1 (24.11)Azure AI Foundry––2.006.003.00131,072aggregatorlitellm
Mistral Large 2.1 (24.11)Google Vertex AI––2.006.003.00131,072aggregatorlitellm
Mistral Medium 3.5Azure AI Foundry––1.507.503.00262,144aggregatorlitellm
Nano Banana 2.1 2 sourcesGoogle––1.507.503.0065,536officialofficial-google
Nano Banana 2.1OpenRouter––1.507.503.0065,536apiopenrouter
Mistral Medium 3.5Mistral AI–eu1.658.253.30262,144apiopenrouter
GPT-5.1OpenRouter––1.2510.003.44400,000apiopenrouter
GPT-5OpenRouter––1.2510.003.44400,000apiopenrouter
Gemini 2.5 ProOpenRouter––1.2510.003.441,048,576apiopenrouter
GPT-5.1 3 sourcesAzure AI Foundry––1.2510.003.44400,000apiopenrouter
GPT-5.1 3 sourcesOpenAI––1.2510.003.44400,000apiopenrouter
GPT-5 3 sourcesAzure AI Foundry––1.2510.003.44400,000apiopenrouter
GPT-5 3 sourcesOpenAI––1.2510.003.44400,000apiopenrouter
Gemini 2.5 Pro 4 sourcesGoogle––1.2510.003.441,048,576officialofficial-google
Gemini 2.5 Pro 2 sourcesDeepInfra––1.2510.003.441,048,576officialofficial-deepinfra
Gemini 2.5 Pro 2 sourcesGoogle Vertex AI––1.2510.003.441,048,576aggregatorlitellm
GPT-5.1Perplexity––1.2510.003.44400,000aggregatorlitellm
GPT-5Perplexity––1.2510.003.44400,000aggregatorlitellm
GPT-5.1Azure AI Foundry–global1.2510.003.44400,000aggregatorlitellm
o3OpenRouter––2.008.003.50200,000apiopenrouter
o3 3 sourcesOpenAI––2.008.003.50200,000apiopenrouter
o3 2 sourcesAzure AI Foundry––2.008.003.50200,000aggregatorlitellm
Qwen3-Max 2 sourcesNovita AI––2.118.453.70262,144aggregatorlitellm
GPT-5.1Azure AI Foundry–swedencentral1.3811.003.78400,000apiopenrouter
GPT-5Azure AI Foundry–swedencentral1.3811.003.78400,000apiopenrouter
GPT-5.1Azure AI Foundry–eu1.3811.003.78400,000aggregatorlitellm
GPT-5Azure AI Foundry–eu1.3811.003.78400,000aggregatorlitellm
GPT-5Azure AI Foundry–us1.3811.003.78400,000aggregatorlitellm
GPT-5.1Azure AI Foundry–us1.3811.003.78400,000aggregatorlitellm
o3Azure AI Foundry–eu2.208.803.85200,000aggregatorlitellm
o3Azure AI Foundry–us2.208.803.85200,000aggregatorlitellm
GPT-6.1 SolOpenRouter––2.0010.004.001,050,000apiopenrouter
Claude Sonnet 5.5OpenRouter––2.0010.004.001,000,000apiopenrouter
GPT-6.1 Sol 3 sourcesAzure AI Foundry––2.0010.004.001,050,000apiopenrouter
GPT-6.1 Sol 4 sourcesOpenAI––2.0010.004.001,050,000officialofficial-openai
Claude Sonnet 5.5Google Vertex AI–global2.0010.004.001,000,000apiopenrouter
Claude Sonnet 5.5 3 sourcesAWS Bedrock––2.0010.004.001,000,000apiopenrouter
Claude Sonnet 5.5Azure AI Foundry–global2.0010.004.001,000,000apiopenrouter
Claude Sonnet 5.5Claude on AWS––2.0010.004.001,000,000apiopenrouter
Claude Sonnet 5.5 4 sourcesAnthropic––2.0010.004.001,000,000officialofficial-anthropic
DeepSeek V3.2 (reasoner)Novita AI––4.004.004.00128,000aggregatorlitellm
Claude Sonnet 5.5 2 sourcesAzure AI Foundry––2.0010.004.001,000,000aggregatorlitellm
Claude Sonnet 5.5Perplexity––2.0010.004.001,000,000aggregatorlitellm
GPT-6.1 SolPerplexity––2.0010.004.001,050,000aggregatorlitellm
Claude Sonnet 5.5 2 sourcesGoogle Vertex AI––2.0010.004.001,000,000aggregatorlitellm
GPT-6.1 Sol 2 sourcesAWS Bedrock–global2.0010.004.001,050,000apiopenrouter
Claude Sonnet 5.5AWS Bedrock–global2.0010.004.001,000,000aggregatorlitellm
Claude Sonnet 5.5DeepInfra––2.0010.004.001,000,000officialofficial-deepinfra
Command A 2 sourcesCohere––2.5010.004.38256,000aggregatorlitellm
GPT-6.1 SolAWS Bedrock–us-east-12.2011.004.401,050,000apiopenrouter

Confidence: official = vendor pricing page · api = OpenRouter per-host · aggregator = community list, may lag · unreviewed = automatic extraction awaiting review. When sources disagree the most authoritative wins. Details: /methodology#confidence