--- title: "LlamaFactory vs CodeGen" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/hiyouga-llamafactory-vs-salesforce-codegen" tools: ["hiyouga-llamafactory", "salesforce-codegen"] --- # LlamaFactory vs CodeGen *GraphCanon updated Jul 12, 2026* ## Verdict Pick LlamaFactory if llamaFactory is a sophisticated tool for fine-tuning numerous large language models and visual language models efficiently using various methods such as LoRA, QLoRA, RLHF, and quantization; pick CodeGen if codeGen is a series of open-source large language models designed for program synthesis. Trained on TPUs, CodeGen offers several versions with varying capabilities from basic code generation to. [LlamaFactory](https://llamafactory.readthedocs.io) reports 73k GitHub stars, 8.9k forks, and 1.1k open issues, last pushed Jul 10, 2026. [CodeGen](https://github.com/salesforce/CodeGen) has 5.2k stars, 423 forks, and 48 open issues, last pushed Jun 2, 2026. Figures are from public GitHub metadata via [LlamaFactory's repository](https://github.com/hiyouga/LlamaFactory) and [CodeGen's repository](https://github.com/salesforce/CodeGen). | | [LlamaFactory](/tools/hiyouga-llamafactory.md) | [CodeGen](/tools/salesforce-codegen.md) | | --- | --- | --- | | Tagline | Unified Efficient Fine-Tuning of 100+ LLMs & VLMs | Family of open-source models for program synthesis. | | Stars | 73,157 | 5,177 | | Forks | 8,937 | 423 | | Open issues | 1,067 | 48 | | Language | Python | Python | | Adopt for | LlamaFactory is a sophisticated tool for fine-tuning numerous large language models and visual language models efficiently using various methods such as LoRA, QLoRA, RLHF, and quantization. | CodeGen is a series of open-source large language models designed for program synthesis. Trained on TPUs, CodeGen offers several versions with varying capabilities from basic code generation to advanced infill sampling. | | Persona | - | - | | Runtime | - | - | | License | Apache-2.0 | Apache-2.0 | | Categories | LLM Frameworks, Model Training | LLM Frameworks, Model Training | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [LlamaFactory](/tools/hiyouga-llamafactory.md) | [CodeGen](/tools/salesforce-codegen.md) | | --- | --- | --- | | Maintenance | Very active (96%) | Steady (60%) | | Days since push | 0d | 39d | | Open issues (now) | 1.1k | 48 | | Owner type | User | Organization | | Full report | [trust report](/tools/hiyouga-llamafactory/trust.md) | [trust report](/tools/salesforce-codegen/trust.md) | ## Decision facts: LlamaFactory - **Adopt for:** LlamaFactory is a sophisticated tool for fine-tuning numerous large language models and visual language models efficiently using various methods such as LoRA, QLoRA, RLHF, and quantization. ## Decision facts: CodeGen - **Adopt for:** CodeGen is a series of open-source large language models designed for program synthesis. Trained on TPUs, CodeGen offers several versions with varying capabilities from basic code generation to advanced infill sampling. ## Choose when ### Choose LlamaFactory if… - Tags unique to LlamaFactory: agent, ai, deepseek, fine-tuning. - When you need to fine-tune over 100 different LLMs or VLMs with efficient methods like LoRA or QLoRA. - More GitHub stars (73k vs 5.2k) - visibility, not fit. ### Choose CodeGen if… - Tags unique to CodeGen: codex, generativemodel, languagemodel, llm. - When you require high-performance model training and code generation that matches or exceeds the performance of OpenAI Codex on specific tasks - Leaner open-issue backlog (48). ## When NOT to use LlamaFactory - When you are looking to fine-tune less popular or niche models that are not supported within the 100+ models covered by LlamaFactory. - If your project specifically requires custom fine-tuning methods not available in this repository, such as certain versions of PEFT (Parameter Efficient Fine-Tuning) techniques excluding LoRA and QLoa ## When NOT to use CodeGen - In scenarios where the model's primary use is not centered around code generation or program synthesis, as its specialized training may limit its effectiveness for other types of generative tasks - If your project strictly requires a smaller memory footprint or simpler deployment because advanced models like CodeGen2.5 require significant computational resources and setup ## Common questions ### What is the difference between LlamaFactory and CodeGen? LlamaFactory: Unified Efficient Fine-Tuning of 100+ LLMs & VLMs. CodeGen: Family of open-source models for program synthesis.. See the comparison table for live GitHub stats and shared categories. ### When should I choose LlamaFactory over CodeGen? Choose LlamaFactory over CodeGen when Tags unique to LlamaFactory: agent, ai, deepseek, fine-tuning; When you need to fine-tune over 100 different LLMs or VLMs with efficient methods like LoRA or QLoRA; More GitHub stars (73k vs 5.2k) - visibility, not fit. ### When should I choose CodeGen over LlamaFactory? Choose CodeGen over LlamaFactory when Tags unique to CodeGen: codex, generativemodel, languagemodel, llm; When you require high-performance model training and code generation that matches or exceeds the performance of OpenAI Codex on specific tasks; Leaner open-issue backlog (48). ### When should I avoid LlamaFactory? When you are looking to fine-tune less popular or niche models that are not supported within the 100+ models covered by LlamaFactory. If your project specifically requires custom fine-tuning methods not available in this repository, such as certain versions of PEFT (Parameter Efficient Fine-Tuning) techniques excluding LoRA and QLoa ### When should I avoid CodeGen? In scenarios where the model's primary use is not centered around code generation or program synthesis, as its specialized training may limit its effectiveness for other types of generative tasks If your project strictly requires a smaller memory footprint or simpler deployment because advanced models like CodeGen2.5 require significant computational resources and setup ### Is LlamaFactory or CodeGen more popular on GitHub? LlamaFactory has more GitHub stars (73,157 vs 5,177). Stars measure visibility, not whether either tool fits your constraints. ### Are LlamaFactory and CodeGen open source? Yes - both are open-source projects on GitHub (LlamaFactory: Apache-2.0, CodeGen: Apache-2.0). ### Where can I find alternatives to LlamaFactory or CodeGen? GraphCanon lists graph-backed alternatives at [LlamaFactory alternatives](/tools/hiyouga-llamafactory/alternatives) and [CodeGen alternatives](/tools/salesforce-codegen/alternatives) ([LlamaFactory markdown twin](/tools/hiyouga-llamafactory/alternatives.md), [CodeGen markdown twin](/tools/salesforce-codegen/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/hiyouga-llamafactory-vs-salesforce-codegen.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, LlamaFactory or CodeGen? LlamaFactory: Very active. CodeGen: Steady. 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 LlamaFactory and CodeGen? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [LlamaFactory trust report](/tools/hiyouga-llamafactory/trust); [CodeGen trust report](/tools/salesforce-codegen/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=hiyouga-llamafactory`](/api/graphcanon/graph?tool=hiyouga-llamafactory) - 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/_