Chitu
Tsinghua PACMAN LabChitu is a production-grade inference engine from Tsinghua's PACMAN group focused on running large models across many kinds of hardware, from pure CPU through single GPU to multi-node clusters with prefill-decode separation. It serves DeepSeek, Qwen, GLM and Kimi models on NVIDIA GPUs and on Chinese accelerators including Ascend 910B, Muxi, Hygon and Moore Threads, with FP8 and FP4 weights converted online to run on hardware without native support. A single chitu.run executable launches multi-node and multi-instance deployments.
Default branch is public-main. README is Chinese-primary with an English copy under docs/en/.
Openness
5 high confidence- license
- Apache-2.0(OSI
- source
- public
- core features withheld
- no — paid professional services offered by email are support beside the published engine, not a withheld component
The LICENSE body is the unmodified Apache-2.0 text, apart from the appendix copyright line naming Qingcheng.AI. The repository is public and active and is the engine itself; the README's paid professional-services offer is support sold beside it, not a gate on the core.
- https://api.github.com/repos/thu-pacman/chitu recorded 2026-09-01
Repo metadata for thu-pacman/chitu - archived false, license Apache-2.0, 2,997 stars, default branch public-main, pushed 2026-09-01.
- https://raw.githubusercontent.com/thu-pacman/chitu/public-main/LICENSE recorded 2026-10-07
LICENSE body, read in full and diffed against the canonical Apache-2.0 text: identical except the appendix copyright line reads "Copyright Qingcheng.AI". No appended terms.
- https://raw.githubusercontent.com/thu-pacman/chitu/public-main/README.md recorded 2026-10-07
README (Chinese), read for a paid tier (professional services by email only - no gated build): "生产级大模型推理引擎" (production-grade LLM inference engine); milestones - DeepSeek-R1 671B single-GPU via CPU+GPU heterogeneous inference, FP4-to-FP8/BF16 online conversion, Ascend 910B native support, Moore Threads adaptation, chitu.run single-file multi-node launcher; licence section states Apache-2.0 with SPDX-marked third-party snippets under LICENSES/ and submodules under third_party/.
Adoption
2 low confidenceChitu is distributed through the repository and its release executables rather than a package registry, leaving GitHub stars as the only adoption signal. The README credits production deployments at China Telecom and other Chinese enterprises, which a star count is likely to understate.
- https://api.github.com/repos/thu-pacman/chitu recorded 2026-09-01
Repo metadata - stargazers_count = 2,997 - for thu-pacman/chitu.
Capability
4 medium confidenceChitu is a production inference engine distinguished by broad support for Chinese accelerator hardware: five vendors plus pure CPU, along with prefill-decode separated cluster serving. Its feature set is comparable to xLLM, RTP-LLM and FastDeploy, while vLLM and SGLang remain ahead on community-wide throughput.
- https://raw.githubusercontent.com/thu-pacman/chitu/public-main/README.md recorded 2026-10-07
Milestones and positioning: v0.6.0 chitu.run single-file multi-node/multi-instance/PD- separation launcher; v0.5.x cluster performance and Moore Threads support; v0.4.0 Ascend/NVIDIA/Muxi/Hygon adaptation serving DeepSeek/Qwen/GLM/Kimi; v0.2.2 single-GPU DeepSeek-R1 671B via CPU+GPU heterogeneous inference; v0.3.0 FP4-to-FP8/BF16 kernels.
Verified 2026-09-27