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 ↑?ContextConfidence?
Amazon Nova Micro 2 sourcesAmazonAWS Bedrock––0.03500.1400.0613128,000aggregator
Amazon Nova MicroAmazonAmazon––0.03500.1400.0613128,000aggregator
Amazon Nova MicroAmazonAWS Bedrock–us0.03500.1400.0613128,000aggregator
Amazon Nova MicroAmazonAWS Bedrock–apac0.03700.1480.0648128,000aggregator
GPT-5 nanoOpenAIOpenAIflex–0.02500.2000.0688400,000api
Phi-4MicrosoftDeepInfraflex–0.05600.1120.070016,384api
Amazon Nova MicroAmazonAWS Bedrock–us-gov-west-10.04200.1680.0735128,000aggregator
Amazon Nova MicroAmazonAWS Bedrock–eu0.04600.1840.0805128,000aggregator
Phi-4MicrosoftOpenRouter––0.07000.1400.087516,384api
Gemini 2.5 Flash-LiteGoogleGoogleflex–0.05000.2000.08751,048,576api
Phi-4 3 sourcesMicrosoftDeepInfra––0.07000.1400.087516,384official
DeepSeek V4 FlashDeepSeekDeepInfraflex+fp8–0.07200.1440.09001,048,576api
GPT-6 Luna 2 sourcesOpenAIOpenAIflex–0.05000.2500.1001,050,000official
MiniMax-M1MiniMaxFireworks AI––0.1000.1000.1001,000,000aggregator
GPT-6 LunaOpenAIOpenAIbatch–0.05000.2500.1001,050,000official
Amazon Nova Lite 2 sourcesAmazonAWS Bedrock––0.06000.2400.105300,000aggregator
Amazon Nova LiteAmazonAmazon––0.06000.2400.105300,000aggregator
Amazon Nova LiteAmazonAWS Bedrock–us0.06000.2400.105300,000aggregator
Amazon Nova LiteAmazonAWS Bedrock–apac0.06300.2520.110300,000aggregator
DeepSeek V4 FlashDeepSeekOpenRouter––0.09000.1800.1131,048,576api
DeepSeek V4 FlashDeepSeekDeepInfrafp8–0.09000.1800.1131,048,576api
DeepSeek V4 Flash 2 sourcesDeepSeekDeepInfra––0.09000.1800.1131,048,576aggregator
Llama 4 ScoutMetaDeepInfraflex+fp8–0.08000.2400.12010,000,000api
DeepSeek V4 FlashDeepSeekDigitalOcean––0.09800.1960.1231,048,576api
Amazon Nova LiteAmazonAWS Bedrock–us-gov-west-10.07200.2880.126300,000aggregator
DeepSeek V4 Flash 2 sourcesDeepSeekFireworks AI––0.1050.2100.1311,048,576api
Amazon Nova LiteAmazonAWS Bedrock–eu0.07800.3120.137300,000aggregator
GPT-5 nanoOpenAIOpenRouter––0.05000.4000.138400,000api
GPT-5 nano 3 sourcesOpenAIAzure AI Foundry––0.05000.4000.138400,000api
GPT-5 nano 3 sourcesOpenAIOpenAI––0.05000.4000.138400,000api
GPT-5 nanoOpenAIPerplexity––0.05000.4000.138400,000aggregator
Llama 4 ScoutMetaOpenRouter––0.1000.3000.15010,000,000api
DeepSeek V4 FlashDeepSeekRelacefp4–0.05000.4500.1501,048,576api
Llama 4 ScoutMetaDeepInfrafp8–0.1000.3000.15010,000,000api
Mistral Small 3.1 2 sourcesMistral AIAzure AI Foundry––0.1000.3000.150131,072aggregator
Llama 4 Scout 2 sourcesMetaDeepInfra––0.1000.3000.15010,000,000aggregator
Mistral Small 3.1 2 sourcesMistral AIGoogle Vertex AI––0.1000.3000.150131,072aggregator
GPT-5 nanoOpenAIAzure AI Foundry–swedencentral0.05500.4400.151400,000api
GPT-5 nanoOpenAIAzure AI Foundry–eu0.05500.4400.151400,000aggregator
GPT-5 nanoOpenAIAzure AI Foundry–us0.05500.4400.151400,000aggregator
Llama 3.3 70B InstructMetaDeepInfrafp8+turbo–0.1000.3200.155128,000api
Llama 3.3 70B InstructMetaHyperbolic––0.1200.3000.165128,000aggregator
DeepSeek V4 FlashDeepSeekSiliconFlowfp8–0.1300.2800.1681,048,576api
DeepSeek V4 FlashDeepSeekDeepInfrafp8+priority–0.1350.2700.1691,048,576api
DeepSeek V4 FlashDeepSeekVenice promo −30%––0.1380.2750.1721,048,576api
Gemini 2.5 Flash-LiteGoogleOpenRouter––0.1000.4000.1751,048,576api
DeepSeek V4 FlashDeepSeekStreamLake promo −41%fp8–0.1400.2800.1751,048,576api
DeepSeek V4 FlashDeepSeekGMI Cloud promo −35%fp8–0.1400.2800.1751,048,576api
DeepSeek V4 FlashDeepSeekBaidufp8–0.1400.2800.1751,048,576api
DeepSeek V4 FlashDeepSeekAtlas Cloudfp4–0.1400.2800.1751,048,576api

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