{"adoption": {"forks": 205, "observed_at": "2026-08-28T04:07:18.888837+00:00", "stars": 2770}, "canonical_url": "https://ross.abutalabs.com/products/fasterdecoding-medusa", "card": {"archived": false, "artifact_type": "library", "description": "Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads", "domain": ["large-language-models", "machine-learning", "deep-learning"], "enriched": true, "function": ["llm-inference", "machine-learning", "gpu-computing"], "health_score": 22, "homepage": "https://sites.google.com/view/medusa-llm", "language": "Jupyter Notebook", "license": "Apache-2.0", "license_family": "permissive", "maturity": "active", "member_repos": ["FasterDecoding/Medusa"], "name": "FasterDecoding/Medusa", "platform": ["python"], "pushed_at": "2024-06-25T12:23:04+00:00", "repo": "FasterDecoding/Medusa", "stars": 2770, "tags": ["speculative-decoding", "decoding-heads", "llm-acceleration", "tree-attention", "inference-optimization", "gpu", "linux"], "topics": ["llm", "llm-inference"], "urls": [], "use_cases": ["speed up llm text generation", "accelerate inference without a draft model", "make local llm hosting faster", "reduce llm generation latency", "add speculative decoding heads to a fine-tuned model", "faster sampling-based generation"], "what_it_is": "Medusa is a framework that accelerates LLM text generation by adding multiple decoding heads to an existing model, avoiding the need for a separate draft model. It achieves roughly 2-3.6x speedups using tree-based attention and a typical acceptance scheme during decoding.", "when_to_avoid": ["you need high-throughput batched serving rather than batch size 1", "you cannot fine-tune or attach new heads to your model", "you need a turnkey inference server rather than a research framework", "your inference stack has no Medusa integration"], "when_to_choose": ["you want faster LLM generation without deploying a separate draft model", "you serve single-batch (batch size 1) inference for local model hosting", "you want parameter-efficient training that leaves the base model untouched", "you need faster non-greedy (sampling) generation"]}, "data_as_of": "2026-08-30T08:39:29.467469+00:00", "members": [{"path": "/products/fasterdecoding-medusa", "repo": "FasterDecoding/Medusa", "role": "main", "score": 18}], "provenance": {"archived": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "artifact_type": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}, "description": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "domain": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}, "enriched": {"inputs": [], "kind": "computed", "method": "enrichment_status"}, "function": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}, "health_score": {"inputs": ["days_since_push", "days_since_release", "archived"], "kind": "computed", "method": "health_v1"}, "homepage": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "language": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "license": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "license_family": {"inputs": ["license"], "kind": "computed", "method": "license_family"}, "maturity": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}, "member_repos": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "name": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "platform": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}, "pushed_at": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "repo": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "stars": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "tags": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}, "topics": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "urls": {"kind": "observed", "observed_at": "2026-08-28T04:07:18.888837+00:00", "source": "github"}, "use_cases": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}, "what_it_is": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}, "when_to_avoid": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}, "when_to_choose": {"confidence": null, "enriched_at": "2026-08-30T08:16:40.037230+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "7bca66513aee517b25141f538e1694ce095abb4e2ccc5721b8e2929f67916b2b", "fetched_at": "2026-08-28T04:07:18.888837+00:00", "kind": "readme", "missing": false, "url": "https://github.com/FasterDecoding/Medusa"}, {"content_hash": "b568a058798811acdf886ff582ace12c38d31afa2f3bb54b20bfe1ff070fb980", "fetched_at": "2026-08-29T09:56:11.151359+00:00", "kind": "homepage", "missing": false, "url": "https://sites.google.com/view/medusa-llm"}], "taxonomy_version": 1}}, "score": {"components": {"activity": 0, "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": 1088, "days_push": 799, "days_rel": null, "gap_med": null, "n_releases_24m": 0}, "score": 18, "version": 2}, "staleness": {"enrichment_outdated": false, "low_confidence": false, "scrape_days": 9, "stale_scrape": false}}