CodeGeeX
CodeGeeX is an open multilingual code generation model implemented in Mindspore and available via PyTorch.
GraphCanon updated 3w · GitHub synced 3w
Decision brief
CodeGeeX is an open-source multilingual code generation model, originally developed with MindSpore and compatible with PyTorch through DeepSpeed integration.
Good fit when
- When you require support for multilingual code generation and your project has a Python-based infrastructure with CUDA GPU availability.
- For projects looking to integrate advanced AI-driven code completion or generation capabilities, given its compatibility with DeepSpeed for performance optimization on large-scale models.
Avoid when
- If your development environment lacks the necessary dependencies like Python 3.7+, CUDA 11+, PyTorch 1.10+, and DeepSpeed 0.6+.
- In scenarios where an open-source solution is not preferable or when support for exclusively one language's syntax is sufficient.
Observed Jul 17, 2026 · Source: enrich:decision_facts
Verify the decision
Maintenance and security
Full trust report- Maintenance
- Dormant (719d since push)
- As of 3w
- Provenance
- Not a fork · Organization account
- As of 3w
- Security (OSV)
- 85 low (85 low)
- As of 1mo
Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.
Install
pip install CodeGeeXSimilar tools
Same-category neighbours. No typed graph edges are catalogued for this tool yet.
Evidence and technical details
Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.
Overview
An open-source project for generating code through a multilingual code-generation model, initially developed with MindSpore and now compatible with Torch. It requires Python 3.7+, CUDA 11+, PyTorch 1.10+, DeepSpeed 0.6+. Users can set up the environment directly or via Docker.
Capability facts
- Languages
- python
Source: github.language · Aug 2, 2026
Categories
Compatibility
Sourced claims from the README excerpt - not unsourced marketing copy.
Source: README excerpt (regex_v1, Aug 2, 2026)
Python 3.7+ / CUDA 11+ / PyTorch 1.10+ / DeepSpeed 0.6+ are required. Install ``codegeex``Source link
Tags
README
Getting Started
CodeGeeX is initially implemented in Mindspore and trained Ascend 910 AI Processors. We provide a torch-compatible version based on Megatron-LM to facilitate usage on GPU platforms.
Installation
Python 3.7+ / CUDA 11+ / PyTorch 1.10+ / DeepSpeed 0.6+ are required. Install codegeex package via:
git clone git@github.com:THUDM/CodeGeeX.git
cd CodeGeeX
pip install -e .
Or use CodeGeeX docker to quickly set up the environment (with nvidia-docker installed):
docker pull codegeex/codegeex:latest
---
## License
Our code is licensed under the [Apache-2.0 license](LICENSE).
Our model is licensed under the [license](MODEL_LICENSE).
For agents
This page has a .md twin and JSON over the API.