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

ModelHostVariantInputOutputBlended ↑?Cache readCache writeBatch inBatch outContextLast changeConfidence?
accounts/fireworks/models/flux-1-dev-controlnet-unionFireworks AI–0.0010000.0010000.001000––––4,096–aggregator
juggernaut-lightning-fluxTogether AI––0.0017000.001700––––––official
sd-xlTogether AI––0.0019000.001900––––––official
p-image-ideogramTogether AI––0.0022500.002250––––––official
juggernaut-pro-fluxTogether AI––0.0049000.004900––––––official
Qwen ImageTogether AI––0.0058000.005800––––––official
Qwen3-Embedding-8BNebius AI Studio–0.0100free0.007500––––40,960–aggregator
BGE M3Novita AI–0.0100free0.007500––––8,192–aggregator
BGE Reranker V2 M3Novita AI–0.0100free0.007500––––8,000–aggregator
Qwen/Qwen2.5-Coder-7BNebius AI Studio–0.01000.03000.0150––––32,768–aggregator
text-embedding-3-smallAzure AI Foundry–0.0200free0.0150––––8,191–aggregator
text-embedding-3-smallOpenAI–0.0200free0.0150––––8,191–aggregator
flux-2-devTogether AI––0.01540.0154––––––official
eu.meta.llama3-2-3b-instruct-v1:0Lambda discontinued–0.01500.02500.0175––––128,000–aggregator
bytedance-seedream-3-0Together AI––0.01800.0180––––––official
Llama 3.2 3B InstructDeepInfra–0.02000.02000.0200––––131,0722025-10-04 ↕aggregator
PaddleOCR VL 2 sourcesNovita AI–0.02000.02000.0200––––16,3842026-01-13 ↓aggregator
Llama 3.2 1B Instruct 2 sourcesNovita AI–0.02000.02000.0200––––60,000–aggregator
Qwen/Qwen2-1.5B-InstructTogether AI–0.02000.02000.0200––––32,768–aggregator
Rerank 5 FastCohere–0.0200–0.0200––––––official
Mistral NemoDekaLLMfp80.01800.03000.0210––––131,072–api
gpt-oss-20bPrime Intellectfp40.02000.03000.02250.0180–––131,072–api
Ling 3.0 Tinyant-digitalfp80.01500.04500.02250.003000–––262,144–api
Ling 3.0 TinyOpenRouter–0.01500.04500.02250.003000–––262,144–api
Mistral Nemoio.net promo −28%–0.02300.03000.02480.0150–––131,0722026-10-06 ↓api
Llama 3.1 8B InstructOpenRouter–0.02000.04000.0250––––131,072–api
Llama 3.1 8B InstructDeepInfrafp80.02000.04000.0250––––131,072–api
Meta-Llama-3.1-8B-Instruct-Turbo 2 sourcesDeepInfra–0.02000.04000.0250––––131,0722026-08-27 ↑official
Llama 3.1 8B Instruct 2 sourcesNovita AI–0.02000.05000.0275––––131,0722026-01-13 ↓aggregator
Llama 3.1 8B InstructLambda discontinued–0.02500.04000.0288––––128,000–aggregator
Mistral NemoOpenRouter–0.02900.03000.0293––––131,072–api
Mistral NemoDeepInfrafp80.02900.03000.0293––––131,0722026-10-09 ↑api
Mistral-Nemo-Instruct-2407 2 sourcesDeepInfra–0.02900.03000.0293––––131,0722026-10-09 ↑official
Mistral NemoParasailfp80.03000.03000.0300––––131,072–api
Llama Prompt Guard 2 22M 2 sourcesGroq–0.03000.03000.0300––––512–aggregator
llama-guard-3-8bNebius AI Studio–0.02000.06000.0300––––131,072–aggregator
meta-llama/Meta-Llama-3.1-8B-InstructNebius AI Studio–0.02000.06000.0300––––131,072–aggregator
Qwen/Qwen2-VL-7B-InstructNebius AI Studio–0.02000.06000.0300––––32,768–aggregator
DeepSeek-OCR 2 sourcesNovita AI–0.03000.03000.0300––––8,1922026-01-13 ↓aggregator
Qwen3 4B 2 sourcesNovita AI–0.03000.03000.0300––––128,0002026-01-13 ↓aggregator
DeepSeek OCR 2 2 sourcesNovita AI–0.03000.03000.0300––––8,192–aggregator
wan-2-6-imageTogether AI––0.03000.0300––––––official
flux-2-proTogether AI––0.03000.0300––––––official
flux-2-flexTogether AI––0.03000.0300––––––official
bytedance-seedream-4-0Together AI––0.03000.0300––––––official
GLM-OCRZhipu AI (Z.ai)–0.03000.03000.0300––––––official
ExtractOpenRouter–0.0400free0.0300––––1,000,000–api
SummarizeOpenRouter–0.0400free0.0300––––1,000,000–api
Ling 3.0 Flash VLOpenRouter–0.02100.06160.03120.004200–––131,072–api
Ling 3.0 FlashOpenRouter–0.02100.06300.03150.004200–––262,144–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