{"adoption": {"forks": 138, "observed_at": "2026-08-28T04:03:20.117644+00:00", "stars": 1040}, "canonical_url": "https://ross.abutalabs.com/products/turboquant-pytorch", "card": {"archived": false, "artifact_type": "library", "description": "From-scratch PyTorch implementation of Google's TurboQuant (ICLR 2026) for LLM KV cache compression. 5x compression at 3-bit with 99.5% attention fidelity.", "domain": ["large-language-models", "deep-learning", "machine-learning", "performance"], "enriched": true, "function": ["machine-learning", "deep-learning", "llm-inference", "caching", "compression"], "health_score": 67, "homepage": null, "language": "Python", "license": "MIT", "license_family": "permissive", "maturity": "active", "member_repos": ["tonbistudio/turboquant-pytorch"], "name": "tonbistudio/turboquant-pytorch", "platform": ["python", "windows", "cross-platform"], "pushed_at": "2026-04-23T19:45:18+00:00", "repo": "tonbistudio/turboquant-pytorch", "stars": 1040, "tags": ["kv-cache-compression", "vector-quantization", "pytorch", "quantization", "llm-optimization", "research-reproduction", "gpu", "linux"], "topics": [], "urls": [], "use_cases": ["compress llm kv cache to fit longer contexts in gpu memory", "quantize key-value cache to 3-4 bits in pytorch", "reproduce turboquant paper results", "run long-context llm inference with lower memory usage", "benchmark kv cache compression against attention fidelity", "implement vector quantization for transformer caches"], "what_it_is": "A from-scratch PyTorch implementation of Google's TurboQuant (ICLR 2026) algorithm for compressing LLM key-value caches, including an improved V3 variant that removes the paper's QJL component. It provides quantized KV cache utilities with verified compression/attention-fidelity benchmarks.", "when_to_avoid": ["you need production-ready, drop-in KV cache compression for a serving stack", "you expect the advertised 5x compression at 3-bit to work for generation - tests show 3-bit output is broken", "you use non-NVIDIA GPUs or non-PyTorch frameworks"], "when_to_choose": ["you need KV cache compression for LLM inference in PyTorch", "you want to reproduce or extend the TurboQuant paper", "you need verified, benchmarked quantization with honest results reporting"]}, "data_as_of": "2026-08-30T08:39:29.467469+00:00", "members": [{"path": "/products/turboquant-pytorch", "repo": "tonbistudio/turboquant-pytorch", "role": "main", "score": 50}], "provenance": {"archived": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "artifact_type": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}, "description": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "domain": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}, "enriched": {"inputs": [], "kind": "computed", "method": "enrichment_status"}, "function": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}, "health_score": {"inputs": ["days_since_push", "days_since_release", "archived"], "kind": "computed", "method": "health_v1"}, "homepage": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "language": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "license": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "license_family": {"inputs": ["license"], "kind": "computed", "method": "license_family"}, "maturity": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}, "member_repos": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "name": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "platform": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}, "pushed_at": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "repo": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "stars": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "tags": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}, "topics": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "urls": {"kind": "observed", "observed_at": "2026-08-28T04:03:20.117644+00:00", "source": "github"}, "use_cases": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}, "what_it_is": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}, "when_to_avoid": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}, "when_to_choose": {"confidence": null, "enriched_at": "2026-08-30T07:03:32.575685+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "3e0cea7b84ecdd17ca49c8332f071dbc6c060ea86fe27c5f0c49c901f05e7548", "fetched_at": "2026-08-28T04:03:20.117644+00:00", "kind": "readme", "missing": false, "url": "https://github.com/tonbistudio/turboquant-pytorch"}], "taxonomy_version": 1}}, "score": {"components": {"activity": 78, "longevity": 11, "rhythm": 35}, "computed_at": "2026-09-03T02:20:16.233290+00:00", "flags": ["no_releases", "young"], "formula": "round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)", "inputs": {"age_days": 161, "days_push": 132, "days_rel": null, "gap_med": null, "n_releases_24m": 0}, "score": 50, "version": 2}, "staleness": {"enrichment_outdated": false, "low_confidence": false, "scrape_days": 9, "stale_scrape": false}}