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 ↑?Cache readCache writeBatch inBatch outContextTokenizerConfidence?
Llama 3.3 70B Instruct 2 sourcesMetaDeepInfra––0.2300.4000.273––––128,000llama3aggregator
Grok 4 FastxAIAzure AI Foundry––0.2000.5000.275––––2,000,000–aggregator
Llama 3.3 70B InstructMetaAkashMLfp8–0.2000.5200.2800.100–––128,000llama3api
Llama 4 Scout 3 sourcesMetaNovita AI––0.1800.5900.283––––10,000,000llama3api
Llama 4 ScoutMetaTogether AI––0.1800.5900.283––––10,000,000llama3aggregator
Mistral Small 4Mistral AIMistral AI–eu0.1650.6600.2890.0165–––262,144mistralapi
Mistral Small 4Mistral AIMistral AI–us0.1650.6600.2890.0165–––262,144mistralapi
Llama 3.3 70B InstructMetaParasailfp8–0.2200.5000.2900.110–––128,000llama3api
DeepSeek V4 FlashDeepSeekAzure AI Foundry–us0.2100.5600.2980.0310–––1,048,576deepseekapi
Qwen3-235B-A22B 2 sourcesAlibaba CloudNebius AI Studio––0.2000.6000.3000.200–––131,072qwen2aggregator
Codestral 25.01Mistral AIGoogle Vertex AI––0.2000.6000.300––––262,144mistralaggregator
Llama 4 MaverickMetaDigitalOcean––0.1880.6530.3040.0500–––1,000,000llama3api
Gemini 2.5 Flash-LiteGoogleGooglepriority–0.1800.7200.3150.01800.150––1,048,576geminiapi
DeepSeek V3.2 (chat)DeepSeekDeepSeek––0.2800.4200.3150.0280–––128,000deepseekaggregator
DeepSeek V3.2 (reasoner)DeepSeekDeepSeek––0.2800.4200.3150.0280–––128,000deepseekaggregator
GPT-5 miniOpenAIOpenAIflex–0.1251.000.3440.0125–––400,000o200kapi
Llama 4 Scout 2 sourcesMetaAzure AI Foundry––0.2000.7800.345––––10,000,000llama3aggregator
Llama 4 Maverick 2 sourcesMetaDeepInfra––0.2000.8000.350––––1,000,000llama3aggregator
Qwen3-235B-A22B 2 sourcesAlibaba CloudNovita AI––0.2000.8000.350––––131,072qwen2aggregator
Llama 4 ScoutMetaGoogle Vertex AI–us-east50.2500.7000.363––––10,000,000llama3api
Llama 4 Scout 2 sourcesMetaGoogle Vertex AI––0.2500.7000.363––0.1250.35010,000,000llama3aggregator
DeepSeek V3.2 (chat)DeepSeekDeepInfraflex+fp4–0.2560.7120.370––––128,000deepseekapi
Qwen3-235B-A22BAlibaba CloudFireworks AI––0.2200.8800.385––––131,072qwen2aggregator
GPT-6 Luna 2 sourcesOpenAIOpenAIfast–0.2001.000.4000.02000.250––1,050,000o200kofficial
DeepSeek V3.2 (reasoner)DeepSeekHyperbolic––0.4000.4000.400––––128,000deepseekaggregator
Llama 4 Maverick 2 sourcesMetaNovita AI––0.2700.8500.415––––1,000,000llama3aggregator
Gemini 2.5 FlashGoogleGoogleflex–0.1501.250.4250.01500.0417––1,048,576geminiapi
DeepSeek V3.2 (chat) 2 sourcesDeepSeekDeepInfra––0.2500.9500.4250.130–––128,000deepseekaggregator
Llama 4 Maverick 2 sourcesMetaAzure AI Foundry––0.2501.000.438––––1,000,000llama3aggregator
Codestral 25.01 2 sourcesMistral AIAzure AI Foundry––0.3000.9000.450––––262,144mistralaggregator
Codestral 25.01Mistral AIMistral AI––0.3000.9000.4500.0300–––262,144mistralaggregator
DeepSeek V3.2 (chat)DeepSeekOpenRouter––0.2571.030.450––––128,000deepseekapi
DeepSeek V3.2 (chat)DeepSeekDeepInfrafp4–0.3200.8900.463––––128,000deepseekapi
DeepSeek V4 FlashDeepSeekWafermxfp4–0.6000.1700.4930.200–––1,048,576deepseekapi
DeepSeek V3.2 (chat)DeepSeekStreamLake promo −10%––0.2861.140.500––––128,000deepseekapi
Grok 3 minixAIAzure AI Foundry––0.2501.270.505––––131,072–aggregator
Llama 4 MaverickMetaOpenRouter––0.3501.000.5130.170–––1,000,000llama3api
Llama 4 MaverickMetaParasailfp8–0.3501.000.5130.170–––1,000,000llama3api
Llama 4 MaverickMetaGoogle Vertex AI–us-east50.3501.150.550––––1,000,000llama3api
Llama 4 Maverick 2 sourcesMetaGoogle Vertex AI––0.3501.150.550––0.1750.5751,000,000llama3aggregator
Qwen3-Coder-480B-A35BAlibaba CloudGoogle Vertex AI––0.2201.800.6150.0220–0.1100.900262,144qwen2aggregator
Inkling SmallThinking MachinesOpenRouter––0.4501.200.6380.100–––524,288–api
Inkling SmallThinking MachinesDeepInfrafp8–0.4501.200.6380.100–––524,288–api
Inkling Small 2 sourcesThinking MachinesDeepInfra––0.4501.200.6380.100–––524,288–aggregator
Inkling SmallThinking MachinesDeepInfrafp4–0.4501.200.6380.100–––524,288–api
Llama 3.3 70B Instruct 2 sourcesMetaGroq––0.5900.7900.6400.295–––128,000llama3api
Qwen3-Coder-480B-A35B 2 sourcesAlibaba CloudNovita AI––0.3801.550.673––––262,144qwen2aggregator
GPT-5 miniOpenAIOpenRouter––0.2502.000.6880.0250–––400,000o200kapi
GPT-5 mini 3 sourcesOpenAIAzure AI Foundry––0.2502.000.6880.0300–0.1251.00400,000o200kapi
GPT-5 mini 3 sourcesOpenAIOpenAI––0.2502.000.6880.0250–0.1251.00400,000o200kapi

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