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 ↑?ContextConfidence?
Command R7BCohereCohere––0.03750.1500.0656–official
Gemma 4 26B A4BGoogleDarkbloom––0.04000.1500.0675262,144api
gpt-oss-120bOpenAIDekaLLM––0.03000.1800.0675131,072api
gpt-oss-20b 2 sourcesOpenAINovita AI––0.04000.1500.0675131,072aggregator
DeepSeek R1 0528 Qwen3 8B 2 sourcesDeepSeekNovita AI––0.06000.09000.0675128,000aggregator
Ling 3.1 FlashOpenRouterOpenRouter––0.04500.1350.06751,000,000api
GPT-5 nanoOpenAIOpenAIflex–0.02500.2000.0688400,000api
gemini-31-flash-lite-imageTogether AITogether AI–––0.06900.0690–official
gpt-oss-120bOpenAIAkashMLfp4–0.03000.1870.0693131,072api
nvidia/NVIDIA-Nemotron-Nano-9B-v2NVIDIADeepInfra––0.04000.1600.0700131,072aggregator
Baichuan M2 32B 2 sourcesNovita AINovita AI––0.07000.07000.0700131,072aggregator
Qwen2.5 7B Instruct 2 sourcesAlibaba CloudNovita AI––0.07000.07000.070032,768aggregator
flux-2-maxTogether AITogether AI––0.0700–0.0700–official
Phi-4MicrosoftDeepInfraflex–0.05600.1120.070016,384api
gpt-oss-120b 3 sourcesOpenAIDeepInfra––0.03700.1700.0703131,072api
Llama 3.2 1B InstructMetaOpenRouter––0.02700.2010.070560,000api
Llama 3.2 1B InstructMetaCloudflare Workers AI––0.02700.2010.070560,000api
arcee-ai/trinity-miniTogether AITogether AI––0.04500.1500.0713128,000aggregator
DeepSeek-V4-Flash-0731DeepInfraDeepInfraflex+fp8–0.04800.1440.07201,048,576api
Mistral Nemo 2 sourcesMistral AINovita AI––0.04000.1700.0725131,072aggregator
Amazon Nova MicroAmazonAWS Bedrock–us-gov-west-10.04200.1680.0735128,000aggregator
Gemma 3 12BGoogleOpenRouter––0.05000.1500.0750131,072api
gpt-oss-20bOpenAISiliconFlowfp8–0.04000.1800.0750131,072api
Gemma 3 12B 4 sourcesGoogleDeepInfra––0.05000.1500.0750131,072official
text-embedding-ada-002OpenAIAzure AI Foundry––0.100free0.07508,192aggregator
Embed v3 MultilingualAzure AI FoundryAzure AI Foundry––0.100free0.0750512aggregator
Embed v3 EnglishAzure AI FoundryAzure AI Foundry––0.100free0.0750512aggregator
Mistral EmbedMistral AIMistral AI––0.100free0.07508,000aggregator
text-embedding-ada-002OpenAIOpenAI––0.100free0.07508,192aggregator
qwen-image-20-proTogether AITogether AI–––0.07500.0750–official
Gemma 3 4BGoogleAWS Bedrock–eu-west-20.06000.1200.0750131,072aggregator
Hy-MT2-1.8BOpenRouterOpenRouter––0.04400.1770.07738,192api
Hy-MT2-1.8BOpenRouterTencent Cloudfp8–0.04400.1770.07738,192api
GLM 5.3 FlashZhipu AI (Z.ai)Wafernvfp4us0.04200.1840.07751,048,576api
DeepSeek V4.1 Flash 1017DeepSeekSAIL Researchfp4us0.07000.1000.07751,048,576api
Embed 5 FastCohereCohere––0.0800–0.0800–official
flux-1-kontext-maxTogether AITogether AI––0.0800–0.0800–official
Gemma 3 27BGoogleDeepInfraflex+fp8–0.06400.1280.0800131,072api
Amazon Nova MicroAmazonAWS Bedrock–eu0.04600.1840.0805128,000aggregator
Granite 4.2 8BOpenRouterOpenRouter––0.06000.1500.0825131,072api
Granite 4.2 8BOpenRouterCoreWeave––0.06000.1500.0825131,072api
NVIDIA-Nemotron-3.5-Lightning 2 sourcesNVIDIADeepInfra––0.06000.1600.0850262,144official
DeepSeek V4.1 Flash 1017DeepSeekparticle-aiflexus0.03000.2500.08501,048,576api
Mistral Small 3.2 24BMistral AIDeepInfraflex+fp8–0.06000.1600.0850256,000api
Particle-1particle-aiparticle-aiflexus0.03000.2500.08501,048,576api
Solar Mini 4OpenRouterOpenRouter––0.05000.2000.0875524,288api
Phi-4MicrosoftOpenRouter––0.07000.1400.087516,384api
MythoMax 13BOpenRouterOpenRouter––0.08000.1100.08758,192api
DeepSeek V4 Flash 0731DeepSeekRelacefp4–0.07000.1400.08751,048,576api
Nemotron 3 Nano 30B A3BNVIDIANovita AIfp4–0.05000.2000.0875262,144api

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