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

ModelHostVariantRegionInputOutputBlended ↑?ContextTokenizerConfidence?
Nemotron 3.5 LightningModelRunfp4–0.07000.2000.1031,048,576–api
gpt-oss-120bMancerfp8–0.05500.2500.104131,072o200kapi
Nemotron 3 Nano 30B A3BOpenRouter––0.06000.2400.105262,144–api
Nova Lite 1.0OpenRouter––0.06000.2400.105300,000–api
Nemotron 3 Nano 30B A3BNebius AI Studiofp8–0.06000.2400.105262,144–api
Nova Lite 1.0AWS Bedrock–eu-west-10.06000.2400.105300,000–api
Nova Lite 1.0AWS Bedrock––0.06000.2400.105300,000–api
Amazon Nova Lite 2 sourcesAWS Bedrock––0.06000.2400.105300,000–aggregator
Model Router 2 sourcesAzure AI Foundry––0.140free0.105200,000–aggregator
Amazon Nova LiteAmazon––0.06000.2400.105300,000–aggregator
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3BNebius AI Studio––0.06000.2400.105262,144–aggregator
nvidia/Nemotron-3-Nano-OmniNebius AI Studio––0.06000.2400.105262,144–aggregator
Nemotron 3.5 Lightning 2 sourcesNebius AI Studio––0.06000.2400.1051,048,576–aggregator
nvidia.nemotron-nano-3-30bAWS Bedrock––0.06000.2400.105256,000–aggregator
DeepSeek V4 Flash 0731DeepInfra––0.08000.1800.1051,048,576deepseekaggregator
Nemotron 3 Nano 30B A3BAWS Bedrock––0.06000.2400.105262,144–aggregator
Amazon Nova LiteAWS Bedrock–us0.06000.2400.105300,000–aggregator
NVIDIA Nemotron Nano 9B v2AWS Bedrock–ap-southeast-20.06180.2370.106128,000–aggregator
gpt-oss-safeguard-20bAWS Bedrock–ap-southeast-20.07210.2060.106131,072o200kaggregator
Mistral Small 3.2 24BOpenRouter––0.07500.2000.106256,000mistralapi
Mistral Small 3.2 24BDeepInfrafp8–0.07500.2000.106256,000mistralapi
Mistral-Small-3.2-24B-Instruct-2506 2 sourcesDeepInfra––0.07500.2000.106128,000–official
Granite 4.2 8BDeepInfra––0.06000.2500.108131,072–api
nvidia/NVIDIA-Nemotron-Nano-9B-v2Together AI––0.06000.2500.108131,072–aggregator
nvidia.nemotron-nano-3-30bAWS Bedrock–ap-southeast-20.06180.2470.108256,000–aggregator
GLM 5.3 FlashWafernvfp4–0.06500.2400.1091,048,576–api
Ministral 3 3B 2512Mistral AI–eu0.1100.1100.110131,072mistralapi
LFM3-1.7BLiquid AIfp8–0.08000.2000.110262,144–api
LFM3-1.7BOpenRouter––0.08000.2000.110262,144–api
eu.meta.llama3-2-1b-instruct-v1:0AWS Bedrock–eu-west-10.1100.1100.110128,000–aggregator
Amazon Nova LiteAWS Bedrock–apac0.06300.2520.110300,000–aggregator
Ling 3.0 Flash VL 2 sourcesNovita AI promo −72%––0.07500.2200.111131,072–api
Ling 3.0 Flash Fin 2 sourcesNovita AI promo −44%––0.07500.2200.111262,144–api
Ling 3.0 Flash Sante 2 sourcesNovita AI promo −44%––0.07500.2200.111262,144–api
GLM 5.3 Flashparticle-aiflex+nvfp4–0.06500.2500.1111,048,576–api
DeepSeek V4 FlashOpenRouter––0.09000.1800.1131,048,576deepseekapi
DeepSeek V4 FlashDeepInfrafp8–0.09000.1800.1131,048,576deepseekapi
Qwen3.5-9BSiliconFlowfp8–0.1000.1500.113262,144qwen2api
Qwen3.5-9B 4 sourcesDeepInfra––0.1000.1500.113262,144qwen2official
DeepSeek V4 Flash 2 sourcesDeepInfra––0.09000.1800.1131,048,576deepseekaggregator
Gemini Embedding 001Google––0.150free0.1132,048–aggregator
Gemini Embedding 001Google Vertex AI––0.150free0.1132,048–aggregator
DeepSeek-V4-FlashDeepInfra––0.09000.1800.1131,000,000–official
Gemma 3 12BDeepInfrapriority–0.07500.2250.113131,072geminiapi
Gemma 4 26B A4BAkashMLfp4–0.07000.2400.113262,144gemmaapi
DeepSeek V4 Flash 0731Wafernvfp4us0.05000.3000.1131,048,576deepseekapi
Qwen3.5-FlashOpenRouter––0.06500.2600.1141,000,000qwen2api
Qwen3.5-FlashAlibaba Cloud––0.06500.2600.1141,000,000qwen2api
GPT OSS 20B 2 sourcesGoogle Vertex AI––0.07000.2500.115131,072–aggregator
DeepSeek V4.1 Flash 1017IonStreamflex–0.07750.2330.1161,048,576deepseekapi

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