Hy4-preview-FP8
Support summary
Hy4-preview-FP8 is a checkpoint published by Tencent, built on the HYV4ForCausalLM architecture. Official recipes or cookbooks cover vLLM and SGLang, without pinning a stable release.
- Documented hardware
- NVIDIA / CUDA
- Documented quantization
- mxfp8
- Documented features
- Multi-token prediction, Reasoning parsers, Speculative decoding, Tensor parallelism, and Tool calling
- Configured context
- 1,048,576 tokens
- License
- apache-2.0
Hardware, quantization and features count only statements about this checkpoint; engine-wide capabilities are listed with each release.
Hugging Face model card
- Publisher
- Tencent
- Task
- text generation
- License
- apache-2.0
- Library
- transformers
- Architecture
- HYV4ForCausalLM
- Weight tensor elements
- 803.5B
- Tensor types
- BF16, F32, F8_E4M3, U8
- Quantization
- modelopt
- Configured context
- 1,048,576 tokens
- Weight access
- Public
- Repository created
- 2026-08-27
Model notes and sources
- tencent/Hy4-preview-FP8 · publisher model card
- tencent/Hy4-preview-FP8 · Hugging Face metadata
- tencent/Hy4-preview-FP8 · publisher configuration
- vllm · models/tencent/Hy4-preview.yaml
- sglang · docs/src/snippets/configs/tencent/hy4-preview.jsx
- sglang · docs/cookbook/autoregressive/Tencent/Hy4-Preview.mdx
- Reviewed
Engine support
3 engines
| Engine | Release | Checkpoint support | Evidence scope | Last known supported |
|---|---|---|---|---|
| vLLM | Version not pinnedRecipe / cookbook | Documented recipeSource: vllm · models/tencent/Hy4-preview.yamlReviewed | No supported release indexed | |
Features, dependencies and evidence for vLLM 0.28.0 | ||||
| SGLang | Version not pinnedRecipe / cookbook | Documented recipeSource 1: sglang · docs/cookbook/autoregressive/Tencent/Hy4-Preview.mdxSource 2: sglang · docs/src/snippets/_deployment.jsxSource 3: sglang · docs/src/snippets/configs/tencent/hy4-preview.jsxSource 4: sglang · docs/src/snippets/configs/tencent/hy4-preview-benchmarks.jsxSource 5: sglang · docs/src/snippets/_playground.jsxReviewed | No supported release indexed | |
Features, dependencies and evidence for SGLang 0.5.19 | ||||
| TensorRT-LLM | 1.2.1Newest indexed | No matching evidence curated | No supported release indexed | |
Features, dependencies and evidence for TensorRT-LLM 1.2.1 | ||||