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 outContextLast changeConfidence?
GPT-5 miniOpenAIPerplexity––0.2502.000.6880.0250–––400,000–aggregator
Qwen3-Coder-480B-A35B 2 sourcesAlibaba CloudDeepInfra––0.4001.600.7000.100–––262,1442026-08-27 ↕aggregator
GLM-4.6Zhipu AI (Z.ai)DeepInfraflex+fp4–0.4001.600.7000.0800–––200,000–api
Llama 3.3 70B InstructMetaOpenRouter––0.7100.7100.7100.710–––128,000–api
Llama 3.3 70B InstructMetaCoreWeave––0.7100.7100.7100.710–––128,000–api
Llama 3.3 70B Instruct 2 sourcesMetaAzure AI Foundry––0.7100.7100.710––––128,000–aggregator
Llama 3.3 70B InstructMetaGoogle Vertex AI–us-central10.7200.7200.720––––128,000–api
Llama 3.3 70B Instruct 3 sourcesMetaGoogle Vertex AI––0.7200.7200.720––0.3600.360128,000–api
Llama 3.3 70B Instruct 2 sourcesMetaAWS Bedrock––0.7200.7200.720––0.3600.360128,000–aggregator
Llama 3.3 70B InstructMetaAWS Bedrock–us0.7200.7200.720––0.3600.360128,000–aggregator
Mistral Large 3 (25.12)Mistral AIOpenRouter––0.5001.500.7500.0500–––262,144–api
Mistral Large 3 (25.12)Mistral AIMistral AIzdr–0.5001.500.7500.0500–––262,144–api
Llama 3.3 70B Instruct 2 sourcesMetaSambaNova––0.6001.200.750––––128,000–api
Mistral Large 3 (25.12)Mistral AIAzure AI Foundry––0.5001.500.750––––262,144–aggregator
Mistral Large 3 (25.12) 2 sourcesMistral AIAWS Bedrock––0.5001.500.750––0.2500.750262,144–aggregator
Mistral Large 3 (25.12) 2 sourcesMistral AIMistral AI––0.5001.500.7500.0500–––262,1442026-03-02 ↓aggregator
GPT-5 miniOpenAIAzure AI Foundry–swedencentral0.2752.200.7560.0330–––400,000–api
GPT-5 miniOpenAIAzure AI Foundry–eu0.2752.200.7560.0275–0.1381.10400,000–aggregator
GPT-5 miniOpenAIAzure AI Foundry–us0.2752.200.7560.0275–0.1381.10400,000–aggregator
GLM-4.6Zhipu AI (Z.ai)Venicefp4–0.4301.750.7600.0800–––200,000–api
Mistral Large 3 (25.12)Mistral AIAWS Bedrock–ap-southeast-20.5151.550.773––––262,144–aggregator
Llama 3.3 70B InstructMetaCloudflare Workers AIfp8–0.2932.250.783––––128,000–api
Qwen3-Coder-480B-A35B 2 sourcesAlibaba CloudAWS Bedrock––0.4501.800.788––––262,1442026-09-06 ↑aggregator
Qwen3-Coder-480B-A35BAlibaba CloudFireworks AI––0.4501.800.788––––262,144–aggregator
Qwen3-235B-A22B 2 sourcesAlibaba CloudAlibaba Cloud––0.4551.820.796––––131,072–api
Mistral Medium 3.1Mistral AIMistral AI––0.4002.000.800––––131,072–aggregator
Qwen3-Coder-480B-A35BAlibaba CloudAWS Bedrock–ap-southeast-20.4641.850.811––––262,144–aggregator
Qwen3-Coder-480B-A35BAlibaba CloudAWS Bedrock–ap-southeast-30.4701.870.820––––262,144–aggregator
Mistral Large 3 (25.12)Mistral AIMistral AI–eu0.5501.650.8250.0550–––262,144–api
Gemini 2.5 FlashGoogleOpenRouter––0.3002.500.8500.03000.0833––1,048,576–api
MiniMax-M1MiniMaxMiniMax––0.4002.200.850––––1,000,000–api
Gemini 2.5 Flash 4 sourcesGoogleGoogle––0.3002.500.8500.03000.08330.1501.251,048,576–official
Gemini 2.5 Flash 3 sourcesGoogleGoogle Vertex AI––0.3002.500.8500.03000.08330.1501.251,048,576–api
Gemini 2.5 Flash 2 sourcesGoogleDeepInfra––0.3002.500.850––––1,048,5762025-10-04 ↑official
GLM-4.6Zhipu AI (Z.ai)OpenRouter––0.5002.000.8750.100–––200,000–api
GLM-4.6Zhipu AI (Z.ai)DeepInfrafp4–0.5002.000.8750.100–––200,000–api
Kimi K2Moonshot AIDeepInfra––0.5002.000.875––––131,072–aggregator
GLM-4.6 3 sourcesZhipu AI (Z.ai)DeepInfra––0.5002.000.8750.100–––200,000–official
DeepSeek V3.2 (chat)DeepSeekTogether AI––0.6001.700.875––––128,000–aggregator
Mistral Large 3 (25.12)Mistral AIAWS Bedrock–ap-south-10.5901.760.883––––262,144–aggregator
Mistral Large 3 (25.12)Mistral AIAWS Bedrock–ap-northeast-10.6101.820.913––––262,144–aggregator
Mistral Large 3 (25.12)Mistral AIAWS Bedrock–sa-east-10.6101.820.913––––262,144–aggregator
Llama 4 MaverickMetaSambaNova––0.6301.800.923––––1,000,000–aggregator
Qwen3-Coder-480B-A35BAlibaba CloudAWS Bedrock–ap-south-10.5302.120.928––––262,144–aggregator
Qwen3-Coder-480B-A35BAlibaba CloudAWS Bedrock–sa-east-10.5402.160.945––––262,144–aggregator
Qwen3-Coder-480B-A35BAlibaba CloudAWS Bedrock–ap-northeast-10.5402.180.950––––262,144–aggregator
GLM-4.6 3 sourcesZhipu AI (Z.ai)Novita AI––0.5502.200.9630.110–––200,0002026-01-13 ↓api
MiniMax-M1 3 sourcesMiniMaxNovita AI––0.5502.200.963––––1,000,0002026-01-13 ↓api
DeepSeek V4 ProDeepSeekDeepSeekoff-peak–0.6601.980.9900.0220–––1,048,576–official
GLM-4.6Zhipu AI (Z.ai)Zhipu AI (Z.ai)fp4–0.6002.201.000.110–––200,000–api

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