{"adoption": {"forks": 328, "observed_at": "2026-08-28T04:07:22.551985+00:00", "stars": 2800}, "canonical_url": "https://ross.abutalabs.com/products/nanotron", "card": {"archived": false, "artifact_type": "library", "description": "Minimalistic large language model 3D-parallelism training", "domain": ["large-language-models", "deep-learning", "machine-learning", "gpu-computing"], "enriched": true, "function": ["llm-training", "deep-learning", "machine-learning"], "health_score": 76, "homepage": null, "language": "Python", "license": "Apache-2.0", "license_family": "permissive", "maturity": "active", "member_repos": ["huggingface/nanotron"], "name": "huggingface/nanotron", "platform": ["python"], "pushed_at": "2026-05-26T10:32:37+00:00", "repo": "huggingface/nanotron", "stars": 2800, "tags": ["3d-parallelism", "pretraining", "transformers", "distributed-training", "huggingface", "gpu", "linux"], "topics": [], "urls": [], "use_cases": ["pretrain a Llama-style transformer from scratch", "train LLMs with data, tensor, and pipeline parallelism", "scale model training across multiple GPUs and nodes", "learn how large-scale LLM training works", "run distributed pretraining on custom datasets", "benchmark distributed training performance"], "what_it_is": "Nanotron is a minimalistic Python library from Hugging Face for pretraining large language models with 3D parallelism (data, tensor, and pipeline parallelism). It offers a simple, flexible API for training transformer models on custom datasets at scale.", "when_to_avoid": ["you only need to fine-tune existing models rather than pretrain from scratch", "you want a turnkey training framework with broad model-type support", "your project requires features outside transformer pretraining", "you need long-term commercial support or extensive documentation"], "when_to_choose": ["you need a lightweight, hackable codebase for LLM pretraining research", "you want to combine data, tensor, and pipeline parallelism in one tool", "you are learning distributed training techniques via the Ultrascale Playbook", "you want PyTorch-native training without heavyweight framework abstractions"]}, "data_as_of": "2026-08-30T08:39:29.467469+00:00", "members": [{"path": "/products/nanotron", "repo": "huggingface/nanotron", "role": "main", "score": 56}], "provenance": {"archived": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "artifact_type": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}, "description": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "domain": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}, "enriched": {"inputs": [], "kind": "computed", "method": "enrichment_status"}, "function": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}, "health_score": {"inputs": ["days_since_push", "days_since_release", "archived"], "kind": "computed", "method": "health_v1"}, "homepage": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "language": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "license": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "license_family": {"inputs": ["license"], "kind": "computed", "method": "license_family"}, "maturity": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}, "member_repos": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "name": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "platform": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}, "pushed_at": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "repo": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "stars": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "tags": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}, "topics": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "urls": {"kind": "observed", "observed_at": "2026-08-28T04:07:22.551985+00:00", "source": "github"}, "use_cases": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}, "what_it_is": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}, "when_to_avoid": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}, "when_to_choose": {"confidence": null, "enriched_at": "2026-08-30T08:14:36.064524+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "4401e5fbc8440eb4b14239bfd3e9995db7a28fb0290d4e68aa97d515fcc3fad8", "fetched_at": "2026-08-28T04:07:22.551985+00:00", "kind": "readme", "missing": false, "url": "https://github.com/huggingface/nanotron"}], "taxonomy_version": 1}}, "score": {"components": {"activity": 84, "longevity": 77, "rhythm": 8}, "computed_at": "2026-09-02T17:46:02.011165+00:00", "flags": [], "formula": "round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)", "inputs": {"age_days": 1087, "days_push": 99, "days_rel": null, "gap_med": null, "n_releases_24m": 0}, "score": 56, "version": 2}, "staleness": {"enrichment_outdated": false, "low_confidence": false, "scrape_days": 9, "stale_scrape": false}}