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 ↑?Batch inBatch outContextConfidence?
DeepSeek V4 FlashParasailfp8–0.1400.2800.175––1,048,576api
Gemini 2.5 Flash-Lite 3 sourcesGoogle Vertex AI––0.1000.4000.1750.05000.2001,048,576api
Gemini 2.5 Flash-Lite 4 sourcesGoogle––0.1000.4000.1750.05000.2001,048,576official
Llama 3.3 Nemotron Super 49B v1 2 sourcesDeepInfra––0.1000.4000.175––131,072aggregator
Llama 3.3 Nemotron Super 49B v1Nebius AI Studio––0.1000.4000.175––131,072aggregator
DeepSeek V4 FlashNebius AI Studio––0.1400.2800.175––1,048,576aggregator
DeepSeek V4 Flash 2 sourcesNovita AI––0.1400.2800.175––1,048,576aggregator
Llama 3.3 70B InstructNebius AI Studio––0.1300.4000.198––128,000aggregator
GPT-6 LunaOpenRouter––0.1000.5000.200––1,050,000api
GPT-6 Luna 4 sourcesOpenAI––0.1000.5000.2000.05000.2501,050,000official
GPT-6 Luna 3 sourcesAzure AI Foundry––0.1000.5000.200––1,050,000api
GPT-6 Luna 2 sourcesAWS Bedrock––0.1000.5000.200––1,050,000aggregator
GPT-6 LunaPerplexity––0.1000.5000.200––1,050,000aggregator
GPT-6 Luna 2 sourcesAWS Bedrock–global0.1000.5000.200––1,050,000api
Claude Haiku 5.5 3 sourcesAWS Bedrock––0.1000.5000.200––1,000,000api
Claude Haiku 5.5 4 sourcesAnthropic––0.1000.5000.2000.05000.2501,000,000official
Claude Haiku 5.5Azure AI Foundry–global0.1000.5000.200––1,000,000api
Claude Haiku 5.5Claude on AWS––0.1000.5000.200––1,000,000api
Claude Haiku 5.5Google Vertex AI–global0.1000.5000.200––1,000,000api
Claude Haiku 5.5OpenRouter––0.1000.5000.200––1,000,000api
Claude Haiku 5.5AWS Bedrock–global0.1000.5000.200––1,000,000aggregator
Claude Haiku 5.5Perplexity––0.1000.5000.200––1,000,000aggregator
Claude Haiku 5.5 2 sourcesAzure AI Foundry––0.1000.5000.200––1,000,000aggregator
Claude Haiku 5.5 2 sourcesGoogle Vertex AI––0.1000.5000.2000.05000.2501,000,000aggregator
Llama 3.3 70B Instruct 3 sourcesNovita AI––0.1350.4000.201––128,000api
Phi-4 2 sourcesAzure AI Foundry––0.1250.5000.219––16,384aggregator
GPT-6 Luna 2 sourcesAzure AI Foundry–us0.1100.5500.220––1,050,000api
GPT-6 Luna 2 sourcesAzure AI Foundry–eu0.1100.5500.220––1,050,000api
GPT-6 LunaAWS Bedrock–us-east-10.1100.5500.220––1,050,000api
GPT-6 LunaAWS Bedrock–us0.1100.5500.220––1,050,000aggregator
Claude Haiku 5.5AWS Bedrock–eu-west-10.1100.5500.220––1,000,000api
Claude Haiku 5.5AWS Bedrock–us-east-10.1100.5500.220––1,000,000api
Claude Haiku 5.5Azure AI Foundry–us0.1100.5500.220––1,000,000api
Claude Haiku 5.5Google Vertex AI–europe0.1100.5500.220––1,000,000api
Claude Haiku 5.5Google Vertex AI–us0.1100.5500.220––1,000,000api
Claude Haiku 5.5AWS Bedrock–apac0.1100.5500.220––1,000,000aggregator
Claude Haiku 5.5AWS Bedrock–au0.1100.5500.220––1,000,000aggregator
Claude Haiku 5.5AWS Bedrock–eu0.1100.5500.220––1,000,000aggregator
Claude Haiku 5.5AWS Bedrock–jp0.1100.5500.220––1,000,000aggregator
Claude Haiku 5.5AWS Bedrock–us0.1100.5500.220––1,000,000aggregator
Llama 4 ScoutDeepInfrafp8+priority–0.1500.4500.225––10,000,000api
Claude Haiku 5.5AWS Bedrock–us-gov-east-10.1200.6000.240––1,000,000aggregator
Claude Haiku 5.5AWS Bedrock–us-gov-west-10.1200.6000.240––1,000,000aggregator
Claude Haiku 5.5AWS Bedrock–us-gov0.1200.6000.240––1,000,000aggregator
DeepSeek V4 FlashMancerfp8–0.1750.5000.256––1,048,576api
Mistral Small 4OpenRouter––0.1500.6000.263––262,144api
Mistral Small 4Mistral AIzdr–0.1500.6000.263––262,144api
Mistral Small 4 3 sourcesMistral AI––0.1500.6000.263––262,144api
DeepSeek V4 Flash 2 sourcesAzure AI Foundry––0.1900.5100.270––1,048,576aggregator
Qwen3-235B-A22B 2 sourcesDeepInfra––0.1800.5400.270––131,072aggregator

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