Same open weights, served by 1 provider. Weights: tencent/Hy3.
| Provider | Their model id | Serving | Input | Output | Cached in | Blended 3:1 | Source | |
|---|---|---|---|---|---|---|---|---|
DeepInfra Deprecated, 2026-10-08 | tencent/Hy3 | flexFP40.8× standard | $0.104 | $0.424 | $0.026 | $0.184 | Verified 5h ago · Changed 7h ago | |
DeepInfra Deprecated, 2026-10-08 | tencent/Hy3 | standardFP4 | $0.13 | $0.53 | $0.033 | $0.23 | Verified 5h ago · Changed 7h 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 | $23.6/mo |
Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.