Tencent
Hy4 preview
A next-generation productivity model with significantly enhanced Agent and complex task execution capabilities.
Model facts
- Context window
- 1.02M tokens
- Maximum output
- 64K tokens
- Cheapest paid input
- $0.67 per 1M tokens
- Cheapest paid output
- $2 per 1M tokens
- Open weights
- yes
- Providers
- 8
Capabilities and modalities
reasoning, tool calling, structured output, temperature control, open weights, text.
Local hardware estimate
At 4K context: Q4_K_M 478.4 GiB, Q8_0 841.5 GiB, F16 1522.5 GiB working memory.
- Parameters
- 780.0 billion
- Architecture
- hyv4
- Model license
- apache-2.0
Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.
Observed price history
- 2026-08-28: $0.834 input / $2.5 output per 1M tokens
- 2026-09-01: $0.67 input / $2 output per 1M tokens
API providers
- Vancine — model id hy4-preview; $0.67 input / $2 output per 1M tokens; provider documentation
- Kilo Gateway — model id tencent/hy4-preview; $0.834 input / $2.5 output per 1M tokens; provider documentation
- OpenCode Go — model id hy4-preview; $0.834 input / $2.5 output per 1M tokens; provider documentation
- OpenRouter — model id tencent/hy4-preview; $0.834 input / $2.5 output per 1M tokens; provider documentation
- Tencent Token Plan — model id hy4-preview; $0.834 input / $2.5 output per 1M tokens; provider documentation
- Tencent TokenHub — model id hy4-preview; $0.834 input / $2.5 output per 1M tokens; provider documentation
- CrossModel — model id tencent/hy4-preview; $0.96 input / $2.88 output per 1M tokens; provider documentation
- Arena — model id hunyuan-hy3-preview; not published input / not published output per 1M tokens; provider documentation