--- title: "femtoGPT vs hazm" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/keyvank-femtogpt-vs-roshan-research-hazm" tools: ["keyvank-femtogpt", "roshan-research-hazm"] --- # femtoGPT vs hazm *GraphCanon updated Aug 22, 2026* ## Verdict Pick femtoGPT if a minimalistic GPT-style language model framework in Rust, suitable for both CPU and GPU inference and training via OpenCL; pick hazm if hazm is a Persian NLP Toolkit used for dependency parsing, embeddings, lemmatization, normalization, POS tagging and tokenization in Python. [femtoGPT](https://discord.gg/wTJFaDVn45) reports 935 GitHub stars, 67 forks, and 10 open issues, last pushed Oct 21, 2025. [hazm](https://www.roshan-ai.ir/hazm/) has 1.4k stars, 208 forks, and 8 open issues, last pushed Apr 1, 2026. Figures are from public GitHub metadata via [femtoGPT's repository](https://github.com/keyvank/femtoGPT) and [hazm's repository](https://github.com/roshan-research/hazm). | | [femtoGPT](/tools/keyvank-femtogpt.md) | [hazm](/tools/roshan-research-hazm.md) | | --- | --- | --- | | Tagline | Pure Rust implementation of a minimal Generative Pretrained Transformer | Persian NLP Toolkit for dependency parsing, embeddings, lemmatization, normalization, POS tagging, and tokenization | | Stars | 935 | 1,417 | | Forks | 67 | 208 | | Open issues | 10 | 8 | | Language | Rust | Python | | Adopt for | A minimalistic GPT-style language model framework in Rust, suitable for both CPU and GPU inference and training via OpenCL. | Hazm is a Persian NLP Toolkit used for dependency parsing, embeddings, lemmatization, normalization, POS tagging and tokenization in Python. | | Persona | developer harness | - | | Runtime | - | - | | License | MIT License, permitting any use as long as all copyright and license information are retained. | MIT | | Categories | LLM Frameworks, Model Training | Data & Retrieval, Model Training | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [femtoGPT](/tools/keyvank-femtogpt.md) | [hazm](/tools/roshan-research-hazm.md) | | --- | --- | --- | | Days since push | 290d | 142d | | Open issues (now) | 10 | 8 | | Stars delta | Unknown | +5 (30d) | | Open issues delta | Unknown | +1 (30d) | | Owner type | User | Organization | | Full report | [trust report](/tools/keyvank-femtogpt/trust.md) | [trust report](/tools/roshan-research-hazm/trust.md) | ## Decision facts: femtoGPT - **Requirements:** Requires the Rust toolchain installed on your system.; If targeting GPU usage, correct installation of GPU drivers along with OpenCL runtimes is necessary. - **Adopt for:** A minimalistic GPT-style language model framework in Rust, suitable for both CPU and GPU inference and training via OpenCL. - **License detail:** MIT License, permitting any use as long as all copyright and license information are retained. - **Persona:** developer harness ## Decision facts: hazm - **Adopt for:** Hazm is a Persian NLP Toolkit used for dependency parsing, embeddings, lemmatization, normalization, POS tagging and tokenization in Python. ## Choose when ### Choose femtoGPT if… - femtoGPT is primarily Rust; hazm is Python. - Requirements: Requires the Rust toolchain installed on your system.; If targeting GPU usage, correct installation of GPU drivers along with OpenCL runtimes is necessary.. - Tags unique to femtoGPT: from-scratch, gpt, gpu, machine-learning. - Also covers LLM Frameworks. - When you want a pure Rust implementation that provides an easy-to-understand basis for learning about the inner workings of AI models. ### Choose hazm if… - hazm is primarily Python; femtoGPT is Rust. - Tags unique to hazm: dependency-parser, embeddings, lemmatization, natural-language-processing. - Also covers Data & Retrieval. - When working exclusively with Farsi language texts where comprehensive processing tools like dependency parsing or POS tagging are required. ## When NOT to use femtoGPT - When high performance is required as femtoGPT operates relatively slower compared to optimized models, especially for large-scale training. - If your project strictly needs CUDA-based optimization specific to NVIDIA GPUs, given that femtoGPT leverages OpenCL for GPU support. - In cases where the project demands a fully tested and production-ready model; femtoGPT's architecture correctness is not guaranteed due to possible implementation errors. ## When NOT to use hazm - If the project involves languages other than Persian, as Hazm lacks capabilities for multilingual support beyond Farsi. - When advanced machine learning models requiring extensive training data and resources are necessary; Hazm provides utilities but does not include state-of-the-art models. ## Common questions ### What is the difference between femtoGPT and hazm? femtoGPT: Pure Rust implementation of a minimal Generative Pretrained Transformer. hazm: Persian NLP Toolkit for dependency parsing, embeddings, lemmatization, normalization, POS tagging, and tokenization. See the comparison table for live GitHub stats and shared categories. ### When should I choose femtoGPT over hazm? Choose femtoGPT over hazm when femtoGPT is primarily Rust; hazm is Python; Requirements: Requires the Rust toolchain installed on your system.; If targeting GPU usage, correct installation of GPU drivers along with OpenCL runtimes is necessary.; Tags unique to femtoGPT: from-scratch, gpt, gpu, machine-learning; Also covers LLM Frameworks; When you want a pure Rust implementation that provides an easy-to-understand basis for learning about the inner workings of AI models. ### When should I choose hazm over femtoGPT? Choose hazm over femtoGPT when hazm is primarily Python; femtoGPT is Rust; Tags unique to hazm: dependency-parser, embeddings, lemmatization, natural-language-processing; Also covers Data & Retrieval; When working exclusively with Farsi language texts where comprehensive processing tools like dependency parsing or POS tagging are required. ### When should I avoid femtoGPT? When high performance is required as femtoGPT operates relatively slower compared to optimized models, especially for large-scale training. If your project strictly needs CUDA-based optimization specific to NVIDIA GPUs, given that femtoGPT leverages OpenCL for GPU support. In cases where the project demands a fully tested and production-ready model; femtoGPT's architecture correctness is not guaranteed due to possible implementation errors. ### When should I avoid hazm? If the project involves languages other than Persian, as Hazm lacks capabilities for multilingual support beyond Farsi. When advanced machine learning models requiring extensive training data and resources are necessary; Hazm provides utilities but does not include state-of-the-art models. ### Is femtoGPT or hazm more popular on GitHub? hazm has more GitHub stars (1,417 vs 935). Stars measure visibility, not whether either tool fits your constraints. ### Are femtoGPT and hazm open source? Yes - both are open-source projects on GitHub (femtoGPT: MIT, hazm: MIT). ### Where can I find alternatives to femtoGPT or hazm? GraphCanon lists graph-backed alternatives at [femtoGPT alternatives](/tools/keyvank-femtogpt/alternatives) and [hazm alternatives](/tools/roshan-research-hazm/alternatives) ([femtoGPT markdown twin](/tools/keyvank-femtogpt/alternatives.md), [hazm markdown twin](/tools/roshan-research-hazm/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/keyvank-femtogpt-vs-roshan-research-hazm.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, femtoGPT or hazm? femtoGPT: Slowing. hazm: Slowing. 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 femtoGPT and hazm? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [femtoGPT trust report](/tools/keyvank-femtogpt/trust); [hazm trust report](/tools/roshan-research-hazm/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=keyvank-femtogpt`](/api/graphcanon/graph?tool=keyvank-femtogpt) - 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/_