大模型
github.com ·
@ggcr
DeepSeek 开源其生产推理引擎 DeepInfer
DeepSeek 在 GitHub 开源 DeepInfer，称其为支撑其大规模生产服务的高性能推理框架，本次发布提供离线推理核心能力。支持 DeepSeek-V4.1-Flash，包含优化内核（FlashMLA、DeepGEMM、DeepEP、TileKernels、DeepSelect 及 TileLang 内核）、Attention DP/均衡 CP、MoE EP+EPLB、DSpark 与置信度调度验证、SWA 有界重放、异步调度、批不变推理、确定性采样和结构化输出。硬件支持 NVIDIA SM100 GPU 与华为 Ascend 950 NPU；完整服务所需的 KV cache、VE/Prefill/Decode 分离和跨数据并行副本负载均衡未包含在本次发布中。
AI推荐理由 DeepSeek 官方开源其生产推理引擎，正文给出具体支持模型（DeepSeek-V4.1-Flash）、并行策略（Attention DP、MoE EP+EPLB）、DSpark 置信度调度验证、SWA 有界重放、批不变推理与确定性采样等可验证特性，并明确支持 NVIDIA SM100 与华为 Ascend 950，信息密度高、来源为官方仓库。局限：发布说明自述缺少 KV cache、PD 分离与跨副本负载均衡，尚非完整服务方案；星数与 Trending 排名仅为传播信号，不代表实际效果。
88
Star 21,065
榜单名次 21
正文
# DeepInfer
DeepInfer is the high-performance inference framework that powers DeepSeek's large-scale production services. This open-source release provides offline inference with the core features needed for serving at scale.
Supported models:
- DeepSeek-V4.1-Flash
Supported features:
- Optimized kernels, model and sampler runtime
- Parallelism: Attention DP / balanced CP, MoE EP + EPLB
- DSpark and confidence-scheduled verification
- SWA bounded replay
- Asynchronous scheduling (zero-overhead batch scheduler)
- Batch-invariant inference
- Deterministic sampling
- Structured outputs
Feature details and configuration
- **Optimized kernels:** [FlashMLA](https://github.com/deepseek-ai/FlashMLA), [DeepGEMM](https://github.com/deepseek-ai/DeepGEMM), [DeepEP](https://github.com/deepseek-ai/DeepEP), [TileKernels](https://github.com/deepseek-ai/TileKernels), [DeepSelect](https://github.com/deepseek-ai/DeepSelect), and custom [TileLang kernels](deepinfer/model_executor/layers/tilelang_ops).
- **DSpark and confidence-scheduled verification:** accelerate decoding with draft tokens and use confidence estimates to schedule verification. Set `DEEPINFER_ENABLE_MTP_VARLEN=1`; `DEEPINFER_USE_NEXTN_DECODE` controls the maximum draft length (default: 5).
- **Decoder SWA bounded replay:** reduce decoder prefill work by replaying only a bounded suffix. Set `DEEPINFER_YOCO_PREFILL_TRUNCATION_MODE=compact` and `DEEPINFER_YOCO_PREFILL_KEEP_TOKENS=140` for the default V4.1 settings; the retained suffix must cover the SWA window plus the maximum backtracking length.
- **Asynchronous scheduling:** overlap CPU batch scheduling with device execution. Enabled by default.
- **Batch-invariant inference:** keep model results identical across different batch compositions. Enable with `DEEPINFER_BATCH_INVARIANT=1` (NVIDIA only).
- **Deterministic sampling:** produce identical outputs for identical input tokens and sampling parameters. Set `DEEPINFER_BATCH_INVARIANT=1 DEEPINFER_ENABLE_MTP_VARLEN=0` and pass a fixed `seq_seed` to `SamplingParams` (NVIDIA only).
- **Structured outputs:** constrain generation to a regex, JSON schema, or EBNF grammar. Pass `so_args={"schema_type": ..., "schema": ...}` to `SamplingParams`; see [examples](examples/_structured_output.py).
Supported hardware:
- NVIDIA SM100 GPUs
- Huawei Ascend 950 NPUs
Required for a complete inference service (not included in this release):
- A KV cache system
- VE/Prefill/Decode disaggregation (most required engine functionality is already implemented)
- Load balancing across data-parallel replicas for both prefill and decode
## Getting started
### Prerequisites
- Linux
- GCC >= 13.0
- Python 3.12
- CUDA 13 (NVIDIA only)
- CANN 9.2 (Ascend only)
### Build from source
Before building DeepInfer, please create and activate an empty Python 3.12 environment.
``` bash
# Use `ascend` instead for the Ascend platform
# Keep it exported in the environment where you run DeepInfer
export DEEPINFER_ACCELERATOR=nvidia
./build.sh
```
### Offline inference
The default checkpoint, [`deepseek-ai/DeepSeek-V4.1-Flash-1017`](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017), is
downloaded into the Hugging Face cache on the first run; set `DEEPINFER_HF_MODEL_DIR=/path/to/DeepSeek-V4.1-Flash-1017` to
run from a local copy instead.
``` bash
python examples/offline_inference.py
```
See [examples](examples/README.md) for details.
## Performance benchmarks
| Prefill | Decode |
| :---: | :---: |
| [](doc/images/prefill_pareto.png) | [](doc/images/decode_pareto.png) |
Measured on 4 SM100 GPUs for prefill and 64 for decode — each within one scale-up domain. To reproduce the benchmarks, see [run_sweep.sh](examples/sweep/run_sweep.sh).
## Evaluation benchmarks
See [eval/README.md](eval/README.md).
## License
DeepInfer is released under the Apache License 2.0. See [LICENSE](LICENSE) for details.
## Acknowledgments
DeepInfer was originally built primarily on [vLLM](https://github.com/vllm-project/vllm), with additional inspiration from [SGLang](https://github.com/sgl-project/sglang) and [TensorRT-LLM](https://github.com/NVIDIA/TensorRT-LLM). We thank their developers.
## Citation
```bibtex
@misc{deepinfer2026,
title = {DeepInfer: High-performance, large-scale inference framework for DeepSeek models},
author = {DeepSeek-AI},
year = {2026},
publisher = {GitHub},
howpublished = {\url{https://github.com/deepseek-ai/DeepInfer}},
}
```