--- title: "LLM4Decompile vs CodeGeeX" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/albertan017-llm4decompile-vs-zai-org-codegeex" tools: ["albertan017-llm4decompile", "zai-org-codegeex"] --- # LLM4Decompile vs CodeGeeX *GraphCanon updated Aug 17, 2026* ## Verdict Pick LLM4Decompile if lLM4Decompile uses large language models to reverse engineer binary code into assembly instructions and potentially source code; pick CodeGeeX if codeGeeX is an open-source multilingual code generation model, originally developed with MindSpore and compatible with PyTorch through DeepSpeed integration. [LLM4Decompile](https://aclanthology.org/2024.emnlp-main.203) reports 7.0k GitHub stars, 546 forks, and 46 open issues, last pushed Feb 12, 2026. [CodeGeeX](https://codegeex.cn) has 8.8k stars, 688 forks, and 188 open issues, last pushed Aug 13, 2024. Figures are from public GitHub metadata via [LLM4Decompile's repository](https://github.com/albertan017/LLM4Decompile) and [CodeGeeX's repository](https://github.com/zai-org/CodeGeeX). | | [LLM4Decompile](/tools/albertan017-llm4decompile.md) | [CodeGeeX](/tools/zai-org-codegeex.md) | | --- | --- | --- | | Tagline | Decompiling Binary Code with Large Language Models | CodeGeeX is an open multilingual code generation model implemented in Mindspore and available via PyTorch. | | Stars | 6,965 | 8,809 | | Forks | 546 | 688 | | Open issues | 46 | 188 | | Language | Python | Python | | Adopt for | LLM4Decompile uses large language models to reverse engineer binary code into assembly instructions and potentially source code. | CodeGeeX is an open-source multilingual code generation model, originally developed with MindSpore and compatible with PyTorch through DeepSpeed integration. | | Persona | - | - | | Runtime | - | - | | License | MIT | Apache-2.0 | | Categories | LLM Frameworks | LLM Frameworks, Model Training | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [LLM4Decompile](/tools/albertan017-llm4decompile.md) | [CodeGeeX](/tools/zai-org-codegeex.md) | | --- | --- | --- | | Maintenance | Slowing (36%) | Dormant (18%) | | Days since push | 186d | 719d | | Open issues (now) | 46 | 188 | | Stars delta | +205 (30d) | Unknown | | Open issues delta | 0 (30d) | Unknown | | Owner type | User | Organization | | Full report | [trust report](/tools/albertan017-llm4decompile/trust.md) | [trust report](/tools/zai-org-codegeex/trust.md) | ## Shared compatibility - **Python**: [LLM4Decompile](/tools/albertan017-llm4decompile.md) - Python runtime; [CodeGeeX](/tools/zai-org-codegeex.md) - Python runtime ## Decision facts: LLM4Decompile - **Pricing:** freemium - The tool itself is open-source under the MIT license, but using it effectively may require access to specific large language models that could have associated costs. - **Requirements:** Min 16 GB RAM; Requires a GPU for optimal performance with the specified model. - **Adopt for:** LLM4Decompile uses large language models to reverse engineer binary code into assembly instructions and potentially source code. ## Decision facts: CodeGeeX - **Adopt for:** CodeGeeX is an open-source multilingual code generation model, originally developed with MindSpore and compatible with PyTorch through DeepSpeed integration. ## Choose when ### Choose LLM4Decompile if… - License: LLM4Decompile is MIT, CodeGeeX is Apache-2.0. - Pricing: The tool itself is open-source under the MIT license, but using it effectively may require access to specific large language models that could have associated costs.. - Requirements: Min 16 GB RAM; Requires a GPU for optimal performance with the specified model.. - Tags unique to LLM4Decompile: binary, decompile, large language models, reverse-engineering. - LLM4Decompile ships Docker support for self-hosted deployment. - When you need a tool that leverages advanced language models for decompiling binaries more effectively than traditional methods. ### Choose CodeGeeX if… - License: CodeGeeX is Apache-2.0, LLM4Decompile is MIT. - Tags unique to CodeGeeX: ai programming tools, code generation, pretrained-models. - Also covers Model Training. - When you require support for multilingual code generation and your project has a Python-based infrastructure with CUDA GPU availability. ## When NOT to use LLM4Decompile - Avoid this tool if you require high precision in recreating exact source code, especially for heavily optimized binaries that lose contextual information during compilation. - Do not use LLM4Decompile when working with less common architectures (e.g., RISC-V) unless explicitly supported or tested by the model. ## When NOT to use CodeGeeX - 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. ## Common questions ### What is the difference between LLM4Decompile and CodeGeeX? LLM4Decompile: Decompiling Binary Code with Large Language Models. CodeGeeX: CodeGeeX is an open multilingual code generation model implemented in Mindspore and available via PyTorch.. See the comparison table for live GitHub stats and shared categories. ### When should I choose LLM4Decompile over CodeGeeX? Choose LLM4Decompile over CodeGeeX when License: LLM4Decompile is MIT, CodeGeeX is Apache-2.0; Pricing: The tool itself is open-source under the MIT license, but using it effectively may require access to specific large language models that could have associated costs.; Requirements: Min 16 GB RAM; Requires a GPU for optimal performance with the specified model.; Tags unique to LLM4Decompile: binary, decompile, large language models, reverse-engineering; LLM4Decompile ships Docker support for self-hosted deployment; When you need a tool that leverages advanced language models for decompiling binaries more effectively than traditional methods. ### When should I choose CodeGeeX over LLM4Decompile? Choose CodeGeeX over LLM4Decompile when License: CodeGeeX is Apache-2.0, LLM4Decompile is MIT; Tags unique to CodeGeeX: ai programming tools, code generation, pretrained-models; Also covers Model Training; When you require support for multilingual code generation and your project has a Python-based infrastructure with CUDA GPU availability. ### When should I avoid LLM4Decompile? Avoid this tool if you require high precision in recreating exact source code, especially for heavily optimized binaries that lose contextual information during compilation. Do not use LLM4Decompile when working with less common architectures (e.g., RISC-V) unless explicitly supported or tested by the model. ### When should I avoid CodeGeeX? 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. ### Is LLM4Decompile or CodeGeeX more popular on GitHub? CodeGeeX has more GitHub stars (8,809 vs 6,965). Stars measure visibility, not whether either tool fits your constraints. ### Are LLM4Decompile and CodeGeeX open source? Yes - both are open-source projects on GitHub (LLM4Decompile: MIT, CodeGeeX: Apache-2.0). ### Where can I find alternatives to LLM4Decompile or CodeGeeX? GraphCanon lists graph-backed alternatives at [LLM4Decompile alternatives](/tools/albertan017-llm4decompile/alternatives) and [CodeGeeX alternatives](/tools/zai-org-codegeex/alternatives) ([LLM4Decompile markdown twin](/tools/albertan017-llm4decompile/alternatives.md), [CodeGeeX markdown twin](/tools/zai-org-codegeex/alternatives.md)), ranked by typed relationship edges rather than popularity votes. ### Is there a machine-readable version of this comparison? Yes. The markdown twin at [this comparison](/compare/albertan017-llm4decompile-vs-zai-org-codegeex.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, LLM4Decompile or CodeGeeX? LLM4Decompile: Slowing. CodeGeeX: Dormant. Compare maintenance labels, days since push, and release cadence in the trust section below - stars alone do not measure maintenance. ### Where are the full trust reports for LLM4Decompile and CodeGeeX? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [LLM4Decompile trust report](/tools/albertan017-llm4decompile/trust); [CodeGeeX trust report](/tools/zai-org-codegeex/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=albertan017-llm4decompile`](/api/graphcanon/graph?tool=albertan017-llm4decompile) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_