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
| Model | Host | Variant | Region | Input | Output | Blended ↑? | Cache read | Cache write | Context | Last change | Confidence? |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Amazon Nova Micro 2 sources | AWS Bedrock | – | – | 0.0350 | 0.140 | 0.0613 | 0.008750 | 0.0350 | 128,000 | – | aggregator |
| Amazon Nova Micro | Amazon | – | – | 0.0350 | 0.140 | 0.0613 | – | – | 128,000 | – | aggregator |
| Amazon Nova Micro | AWS Bedrock | – | us | 0.0350 | 0.140 | 0.0613 | 0.008750 | – | 128,000 | – | aggregator |
| Amazon Nova Micro | AWS Bedrock | – | apac | 0.0370 | 0.148 | 0.0648 | 0.009250 | – | 128,000 | – | aggregator |
| GPT-5 nano | OpenAI | flex | – | 0.0250 | 0.200 | 0.0688 | 0.002500 | – | 400,000 | – | api |
| Phi-4 | DeepInfra | flex | – | 0.0560 | 0.112 | 0.0700 | – | – | 16,384 | – | api |
| Amazon Nova Micro | AWS Bedrock | – | us-gov-west-1 | 0.0420 | 0.168 | 0.0735 | 0.0105 | – | 128,000 | – | aggregator |
| Amazon Nova Micro | AWS Bedrock | – | eu | 0.0460 | 0.184 | 0.0805 | 0.0115 | – | 128,000 | – | aggregator |
| Phi-4 | OpenRouter | – | – | 0.0700 | 0.140 | 0.0875 | – | – | 16,384 | – | api |
| Gemini 2.5 Flash-Lite | flex | – | 0.0500 | 0.200 | 0.0875 | 0.005000 | 0.0417 | 1,048,576 | – | api | |
| Phi-4 3 sources | DeepInfra | – | – | 0.0700 | 0.140 | 0.0875 | – | – | 16,384 | – | official |
| DeepSeek V4 Flash | DeepInfra | flex+fp8 | – | 0.0720 | 0.144 | 0.0900 | 0.0144 | – | 1,048,576 | – | api |
| GPT-6 Luna 2 sources | OpenAI | flex | – | 0.0500 | 0.250 | 0.100 | 0.005000 | 0.0625 | 1,050,000 | – | official |
| MiniMax-M1 | Fireworks AI | – | – | 0.100 | 0.100 | 0.100 | – | – | 1,000,000 | – | aggregator |
| GPT-6 Luna | OpenAI | batch | – | 0.0500 | 0.250 | 0.100 | 0.005000 | 0.0625 | 1,050,000 | – | official |
| Amazon Nova Lite 2 sources | AWS Bedrock | – | – | 0.0600 | 0.240 | 0.105 | 0.0150 | 0.0600 | 300,000 | – | aggregator |
| Amazon Nova Lite | Amazon | – | – | 0.0600 | 0.240 | 0.105 | – | – | 300,000 | – | aggregator |
| Amazon Nova Lite | AWS Bedrock | – | us | 0.0600 | 0.240 | 0.105 | 0.0150 | – | 300,000 | – | aggregator |
| Amazon Nova Lite | AWS Bedrock | – | apac | 0.0630 | 0.252 | 0.110 | 0.0158 | – | 300,000 | – | aggregator |
| DeepSeek V4 Flash | OpenRouter | – | – | 0.0900 | 0.180 | 0.113 | 0.0180 | – | 1,048,576 | – | api |
| DeepSeek V4 Flash | DeepInfra | fp8 | – | 0.0900 | 0.180 | 0.113 | 0.0180 | – | 1,048,576 | – | api |
| DeepSeek V4 Flash 2 sources | DeepInfra | – | – | 0.0900 | 0.180 | 0.113 | 0.0180 | 0.113 | 1,048,576 | – | aggregator |
| Llama 4 Scout | DeepInfra | flex+fp8 | – | 0.0800 | 0.240 | 0.120 | – | – | 10,000,000 | – | api |
| DeepSeek V4 Flash | DigitalOcean | – | – | 0.0980 | 0.196 | 0.123 | 0.0196 | – | 1,048,576 | – | api |
| Amazon Nova Lite | AWS Bedrock | – | us-gov-west-1 | 0.0720 | 0.288 | 0.126 | 0.0180 | – | 300,000 | – | aggregator |
| DeepSeek V4 Flash 2 sources | Fireworks AI | – | – | 0.105 | 0.210 | 0.131 | 0.0210 | – | 1,048,576 | – | api |
| Amazon Nova Lite | AWS Bedrock | – | eu | 0.0780 | 0.312 | 0.137 | 0.0195 | – | 300,000 | – | aggregator |
| GPT-5 nano | OpenRouter | – | – | 0.0500 | 0.400 | 0.138 | 0.005000 | – | 400,000 | – | api |
| GPT-5 nano 3 sources | Azure AI Foundry | – | – | 0.0500 | 0.400 | 0.138 | 0.0100 | – | 400,000 | – | api |
| GPT-5 nano 3 sources | OpenAI | – | – | 0.0500 | 0.400 | 0.138 | 0.005000 | – | 400,000 | – | api |
| GPT-5 nano | Perplexity | – | – | 0.0500 | 0.400 | 0.138 | 0.005000 | – | 400,000 | – | aggregator |
| Llama 4 Scout | OpenRouter | – | – | 0.100 | 0.300 | 0.150 | – | – | 10,000,000 | – | api |
| DeepSeek V4 Flash | Relace | fp4 | – | 0.0500 | 0.450 | 0.150 | 0.0500 | – | 1,048,576 | 2026-11-11 ↑ | api |
| Llama 4 Scout | DeepInfra | fp8 | – | 0.100 | 0.300 | 0.150 | – | – | 10,000,000 | – | api |
| Mistral Small 3.1 2 sources | Azure AI Foundry | – | – | 0.100 | 0.300 | 0.150 | – | – | 131,072 | 2026-03-03 ↓ | aggregator |
| Llama 4 Scout 2 sources | DeepInfra | – | – | 0.100 | 0.300 | 0.150 | – | – | 10,000,000 | 2026-08-27 ↑ | aggregator |
| Mistral Small 3.1 2 sources | Google Vertex AI | – | – | 0.100 | 0.300 | 0.150 | – | – | 131,072 | 2026-09-11 ↓ | aggregator |
| GPT-5 nano | Azure AI Foundry | – | swedencentral | 0.0550 | 0.440 | 0.151 | 0.0110 | – | 400,000 | – | api |
| GPT-5 nano | Azure AI Foundry | – | eu | 0.0550 | 0.440 | 0.151 | 0.005500 | – | 400,000 | – | aggregator |
| GPT-5 nano | Azure AI Foundry | – | us | 0.0550 | 0.440 | 0.151 | 0.005500 | – | 400,000 | – | aggregator |
| Llama 3.3 70B Instruct | DeepInfra | fp8+turbo | – | 0.100 | 0.320 | 0.155 | – | – | 128,000 | – | api |
| Llama 3.3 70B Instruct | Hyperbolic | – | – | 0.120 | 0.300 | 0.165 | – | – | 128,000 | – | aggregator |
| DeepSeek V4 Flash | SiliconFlow | fp8 | – | 0.130 | 0.280 | 0.168 | 0.0280 | – | 1,048,576 | – | api |
| DeepSeek V4 Flash | DeepInfra | fp8+priority | – | 0.135 | 0.270 | 0.169 | 0.0270 | – | 1,048,576 | – | api |
| DeepSeek V4 Flash | Venice promo −30% | – | – | 0.138 | 0.275 | 0.172 | 0.0280 | – | 1,048,576 | – | api |
| Gemini 2.5 Flash-Lite | OpenRouter | – | – | 0.100 | 0.400 | 0.175 | 0.0100 | 0.0833 | 1,048,576 | – | api |
| DeepSeek V4 Flash | StreamLake promo −41% | fp8 | – | 0.140 | 0.280 | 0.175 | 0.0280 | – | 1,048,576 | – | api |
| DeepSeek V4 Flash | GMI Cloud promo −35% | fp8 | – | 0.140 | 0.280 | 0.175 | 0.0280 | – | 1,048,576 | – | api |
| DeepSeek V4 Flash | Baidu | fp8 | – | 0.140 | 0.280 | 0.175 | 0.0280 | – | 1,048,576 | – | api |
| DeepSeek V4 Flash | Atlas Cloud | fp4 | – | 0.140 | 0.280 | 0.175 | 0.0280 | – | 1,048,576 | – | 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