Model information
Released
2026-04-20
Context window
256K
Max output
64K
Model parameters
299B
Weights
Open
Modalities
Input
Text
Output
Text
Identity
- Version
- hy3-preview
- Aliases
- Hy3 preview
Resources
Capabilities
Open weightReasoningTemperatureTool call
Providers
| Provider | Provider model ID | Protocol | Context | Max output | Input | Output | Capabilities | Notes | Pricing | Streaming |
|---|---|---|---|---|---|---|---|---|---|---|
| Tencent TokenHub | hy3-preview | openai_chat | 256K | 64K | ReasoningTool callTemperature | - | input $0/M · output $0/M cache read $0/M · cache write $0/M | Yes | ||
| CrossModel | tencent/hy3-preview | openai_chat | 256K | 64K | ReasoningTool callTemperature | - | input $0.19/M · output $0.63/M cache read $0.063/M · cache write $0.19/M | Yes | ||
| NanoGPT | tencent/hy3-preview | openai_chat | 256K | 64K | - | - | - | Yes | ||
| SiliconFlow | tencent/Hy3-preview | openai_chat | 256K | 64K | ReasoningTool callTemperature | - | input $0.066/M · output $0.26/M cache read $0.029/M | Yes | ||
| ZenMux | tencent/hy3-preview | openai_chat | 256K | 64K | ReasoningTool callTemperature | - | input $0.172/M · output $0.572/M cache read $0.058/M · cache write $0/M | Yes | ||
| AIHubMix | hy3-preview | openai_chat | 256K | 64K | ReasoningTool callStructuredTemperature | - | input $0.17/M · output $0.567/M cache read $0.051/M | Yes | ||
| Kilo Gateway | tencent/hy3-preview | openai_chat | 256K | 64K | ReasoningTool callTemperature | - | input $0.18/M · output $0.6/M cache read $0.06/M | Yes | ||
| OpenRouter | tencent/hy3-preview | openai_chat | 256K | 64K | ReasoningTool callTemperature | Reasoning levels: high, low, none, Default reasoning: high, Hugging Face: tencent/Hy3-preview | input $0.063/M · output $0.21/M cache read $0.021/M | Yes | ||
| Tencent Hunyuan | hy3-preview | custom | 256K | 64K | - | - | - | Yes |