Same open weights, served by 2 providers. Weights: deepseek-ai/DeepSeek-V3.1.
| Provider | Their model id | Serving | Input | Output | Cached in | Blended 3:1 | Source | |
|---|---|---|---|---|---|---|---|---|
DeepInfra | deepseek-ai/DeepSeek-V3.1 | flexFP40.8× standard | $0.20 | $0.76 | $0.104 | $0.34 | Verified 7h ago · Changed 9h ago | |
DeepInfra | deepseek-ai/DeepSeek-V3.1 | standardFP4 | $0.25 | $0.95 | $0.13 | $0.425 | Verified 7h ago · Changed 9h ago | |
AWS Bedrock | deepseek.v3-v1:0 | standardQuantization not disclosed | $0.58 | $1.68 | — | $0.855 | Verified 9h ago · Changed 9h ago |
Same weights don't always mean identical outputs: quantization, context limits and serving stacks can change quality, especially for tool calling. Prices are USD per 1M tokens; hyperscaler prices are for the reference region shown.
| DeepInfra fp4 | $44/mo |
| AWS Bedrock | $91.6/mo |
Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.