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 every active offering, including the long tail. Show curated models only

More columns: - Vendor + Region + Cache + Batch + Tokenizer - Last change + Source All columns Default columns

ModelVendorHostVariantRegionInputOutputBlended ↑?ContextLast changeConfidence?
GPT-5 nanoOpenAIOpenRouter––0.05000.4000.138400,000–api
Qwen3.5-9BAlibaba CloudParasail––0.1000.2500.138262,144–api
GPT-5 nano 3 sourcesOpenAIAzure AI Foundry––0.05000.4000.138400,000–api
GPT-5 nano 3 sourcesOpenAIOpenAI––0.05000.4000.138400,000–api
Gemma 4 26B A4B 3 sourcesGoogleDeepInfra––0.07000.3400.138262,144–official
Gemma 4 26B A4BGoogleMakora––0.09000.2800.138262,1442026-11-13 ↑api
GPT-5 nanoOpenAIPerplexity––0.05000.4000.138400,000–aggregator
Gemma 3 27B 3 sourcesGoogleNovita AI––0.1190.2000.139131,0722026-01-13 ↓api
Phi 4 MultimodalMicrosoftAzure AI Foundry––0.08000.3200.140131,0722025-04-14 ↓aggregator
Gemma 3 12B 2 sourcesGoogleAWS Bedrock––0.09000.2900.140131,072–aggregator
Hermes 2 Pro Llama 3 8B 2 sourcesNovita AINovita AI––0.1400.1400.1408,1922026-01-13 ↓aggregator
Phi-4-multimodalMicrosoftAzure AI Foundry––0.08000.3200.140128,000–aggregator
Gemma 4 26B A4BGoogleNextBit promo −15%––0.09000.3000.143262,144–api
Qwen3 30B A3B Instruct 2507Alibaba CloudDekaLLM––0.09000.3000.143262,144–api
Qwen3 30B A3B Instruct 2507Alibaba CloudSiliconFlowfp8–0.09000.3000.143262,144–api
Mistral Small 3.2 24BMistral AIParasail––0.09000.3000.143256,000–api
Llama Guard 4 12BMetaDeepInfraflex–0.1440.1440.144163,840–api
Gemma 3 12BGoogleAWS Bedrock–ap-southeast-20.09270.2990.144131,072–aggregator
Hy3OpenRouterTencent Cloudfp8+off-peak–0.08250.3300.144262,144–api
Hy3OpenRouterOpenRouteroff-peak–0.08250.3300.144262,144–api
MiMo-V2.6-FlashOpenRouterio.netfp8–0.1000.2800.1451,050,0002026-10-23 ↓api
MiMo-V2.6-FlashOpenRouterDarkbloomfp4–0.1000.2800.1451,050,0002026-10-06 ↓api
GLM 4.7 Flash 2 sourcesZhipu AI (Z.ai)DeepInfra––0.06000.4000.145200,000–aggregator
gpt-oss-120bOpenAISAIL Researchfp4–0.06000.4000.145131,072–api
GLM 4.7 FlashZhipu AI (Z.ai)OpenRouter––0.06050.4000.145200,000–api
GLM 4.7 FlashZhipu AI (Z.ai)Cloudflare Workers AI––0.06050.4000.145200,000–api
DeepSeek V4.1 Flash 1017DeepSeekambientfp4–0.09800.2940.1471,048,576–api
Gemma 3 27BGoogleNextBitfp8–0.08000.3500.148131,072–api
gpt-oss-20bOpenAIAWS Bedrock–ap-south-10.08000.3500.148131,072–aggregator
MiMo-V2.5OpenRouterOpenRouter––0.1190.2380.1491,000,000–api
DeepSeek V4 Flash 0731DeepSeekDigitalOcean––0.1190.2380.1491,048,576–api
Ministral 3 8B 2512Mistral AIOpenRouter––0.1500.1500.150262,144–api
Voxtral Small 24B 2507Mistral AIOpenRouter––0.1000.3000.15032,768–api
Qwen3 30B A3B Instruct 2507Alibaba CloudOpenRouter––0.1000.3000.150262,144–api
Qwen3 14BAlibaba CloudOpenRouter––0.1200.2400.15040,960–api
Llama 4 ScoutMetaOpenRouter––0.1000.3000.15010,000,000–api
DeepSeek V4 FlashDeepSeekRelacefp4–0.05000.4500.1501,048,5762026-11-11 ↑api
Step 3.5 FlashOpenRouterSiliconFlowfp8–0.1000.3000.150262,144–api
Ministral 3 8B 2512Mistral AIMistral AIzdr–0.1500.1500.150262,144–api
Ministral 3 8B 2512 3 sourcesMistral AIMistral AI––0.1500.1500.150262,144–api
Voxtral Small 24B 2507Mistral AIMistral AIzdr–0.1000.3000.15032,768–api
Voxtral Small 24B 2507Mistral AIMistral AI––0.1000.3000.15032,768–api
gpt-oss-120bOpenAIDigitalOcean––0.06000.4200.150131,072–api
Qwen3 30B A3B Instruct 2507Alibaba CloudNebius AI Studiofp8–0.1000.3000.150262,144–api
Mistral Small 3.2 24BMistral AIMistral AI–eu0.1000.3000.150256,000–api
Qwen3 14BAlibaba CloudDeepInfrafp8–0.1200.2400.15040,960–api
Llama 4 ScoutMetaDeepInfrafp8–0.1000.3000.15010,000,000–api
Gemma 3 27BGoogleNebius AI Studiofp8–0.1000.3000.150131,072–api
Mistral NemoMistral AIMistral AI–eu0.1500.1500.150131,072–api
Mistral NemoMistral AIAzure AI Foundry––0.1500.1500.150131,072–aggregator

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