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

ModelVendorHostVariantRegionInputOutputBlended ↑?ContextTokenizerLast changeConfidence?Source
DeepSeek V4 FlashDeepSeekParasailfp8–0.1400.2800.1751,048,576deepseek–apiopenrouter
Gemini 2.5 Flash-Lite 3 sourcesGoogleGoogle Vertex AI––0.1000.4000.1751,048,576gemini–apiopenrouter
Gemini 2.5 Flash-Lite 4 sourcesGoogleGoogle––0.1000.4000.1751,048,576gemini–officialofficial-google
Llama 3.3 Nemotron Super 49B v1 2 sourcesNVIDIADeepInfra––0.1000.4000.175131,072llama3–aggregatorlitellm
Llama 3.3 Nemotron Super 49B v1NVIDIANebius AI Studio––0.1000.4000.175131,072llama3–aggregatorlitellm
DeepSeek V4 FlashDeepSeekNebius AI Studio––0.1400.2800.1751,048,576deepseek–aggregatorlitellm
DeepSeek V4 Flash 2 sourcesDeepSeekNovita AI––0.1400.2800.1751,048,576deepseek–aggregatorlitellm
Llama 3.3 70B InstructMetaNebius AI Studio––0.1300.4000.198128,000llama3–aggregatorlitellm
GPT-6 LunaOpenAIOpenRouter––0.1000.5000.2001,050,000o200k–apiopenrouter
GPT-6 Luna 4 sourcesOpenAIOpenAI––0.1000.5000.2001,050,000o200k–officialofficial-openai
GPT-6 Luna 3 sourcesOpenAIAzure AI Foundry––0.1000.5000.2001,050,000o200k–apiopenrouter
GPT-6 Luna 2 sourcesOpenAIAWS Bedrock––0.1000.5000.2001,050,000o200k–aggregatorlitellm
GPT-6 LunaOpenAIPerplexity––0.1000.5000.2001,050,000o200k–aggregatorlitellm
GPT-6 Luna 2 sourcesOpenAIAWS Bedrock–global0.1000.5000.2001,050,000o200k–apiopenrouter
Claude Haiku 5.5 3 sourcesAnthropicAWS Bedrock––0.1000.5000.2001,000,000claude–apiopenrouter
Claude Haiku 5.5 4 sourcesAnthropicAnthropic––0.1000.5000.2001,000,000claude–officialofficial-anthropic
Claude Haiku 5.5AnthropicAzure AI Foundry–global0.1000.5000.2001,000,000claude–apiopenrouter
Claude Haiku 5.5AnthropicClaude on AWS––0.1000.5000.2001,000,000claude–apiopenrouter
Claude Haiku 5.5AnthropicGoogle Vertex AI–global0.1000.5000.2001,000,000claude–apiopenrouter
Claude Haiku 5.5AnthropicOpenRouter––0.1000.5000.2001,000,000claude–apiopenrouter
Claude Haiku 5.5AnthropicAWS Bedrock–global0.1000.5000.2001,000,000claude–aggregatorlitellm
Claude Haiku 5.5AnthropicPerplexity––0.1000.5000.2001,000,000claude–aggregatorlitellm
Claude Haiku 5.5 2 sourcesAnthropicAzure AI Foundry––0.1000.5000.2001,000,000claude–aggregatorlitellm
Claude Haiku 5.5 2 sourcesAnthropicGoogle Vertex AI––0.1000.5000.2001,000,000claude–aggregatorlitellm
Llama 3.3 70B Instruct 3 sourcesMetaNovita AI––0.1350.4000.201128,000llama32026-01-13 ↓apiopenrouter
Phi-4 2 sourcesMicrosoftAzure AI Foundry––0.1250.5000.21916,384––aggregatorlitellm
GPT-6 Luna 2 sourcesOpenAIAzure AI Foundry–us0.1100.5500.2201,050,000o200k–apiopenrouter
GPT-6 Luna 2 sourcesOpenAIAzure AI Foundry–eu0.1100.5500.2201,050,000o200k–apiopenrouter
GPT-6 LunaOpenAIAWS Bedrock–us-east-10.1100.5500.2201,050,000o200k–apiopenrouter
GPT-6 LunaOpenAIAWS Bedrock–us0.1100.5500.2201,050,000o200k–aggregatorlitellm
Claude Haiku 5.5AnthropicAWS Bedrock–eu-west-10.1100.5500.2201,000,000claude–apiopenrouter
Claude Haiku 5.5AnthropicAWS Bedrock–us-east-10.1100.5500.2201,000,000claude–apiopenrouter
Claude Haiku 5.5AnthropicAzure AI Foundry–us0.1100.5500.2201,000,000claude–apiopenrouter
Claude Haiku 5.5AnthropicGoogle Vertex AI–europe0.1100.5500.2201,000,000claude–apiopenrouter
Claude Haiku 5.5AnthropicGoogle Vertex AI–us0.1100.5500.2201,000,000claude–apiopenrouter
Claude Haiku 5.5AnthropicAWS Bedrock–apac0.1100.5500.2201,000,000claude–aggregatorlitellm
Claude Haiku 5.5AnthropicAWS Bedrock–au0.1100.5500.2201,000,000claude–aggregatorlitellm
Claude Haiku 5.5AnthropicAWS Bedrock–eu0.1100.5500.2201,000,000claude–aggregatorlitellm
Claude Haiku 5.5AnthropicAWS Bedrock–jp0.1100.5500.2201,000,000claude–aggregatorlitellm
Claude Haiku 5.5AnthropicAWS Bedrock–us0.1100.5500.2201,000,000claude–aggregatorlitellm
Llama 4 ScoutMetaDeepInfrafp8+priority–0.1500.4500.22510,000,000llama3–apiopenrouter
Claude Haiku 5.5AnthropicAWS Bedrock–us-gov-east-10.1200.6000.2401,000,000claude–aggregatorlitellm
Claude Haiku 5.5AnthropicAWS Bedrock–us-gov-west-10.1200.6000.2401,000,000claude–aggregatorlitellm
Claude Haiku 5.5AnthropicAWS Bedrock–us-gov0.1200.6000.2401,000,000claude–aggregatorlitellm
DeepSeek V4 FlashDeepSeekMancerfp8–0.1750.5000.2561,048,576deepseek2026-11-07 ↓apiopenrouter
Mistral Small 4Mistral AIOpenRouter––0.1500.6000.263262,144mistral–apiopenrouter
Mistral Small 4Mistral AIMistral AIzdr–0.1500.6000.263262,144mistral–apiopenrouter
Mistral Small 4 3 sourcesMistral AIMistral AI––0.1500.6000.263262,144mistral2026-08-21 ↑apiopenrouter
DeepSeek V4 Flash 2 sourcesDeepSeekAzure AI Foundry––0.1900.5100.2701,048,576deepseek–aggregatorlitellm
Qwen3-235B-A22B 2 sourcesAlibaba CloudDeepInfra––0.1800.5400.270131,072qwen22026-08-27 ↓aggregatorlitellm

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