{"adoption": {"forks": 115, "observed_at": "2026-08-28T04:04:29.043188+00:00", "stars": 1354}, "canonical_url": "https://ross.abutalabs.com/products/bytedance-flux", "card": {"archived": false, "artifact_type": "library", "description": "A fast communication-overlapping library for tensor/expert parallelism on GPUs.", "domain": ["deep-learning", "large-language-models", "gpu-computing", "microservices", "machine-learning"], "enriched": true, "function": ["gpu-computing", "machine-learning", "llm-training", "concurrency"], "health_score": 52, "homepage": null, "language": "C++", "license": "Apache-2.0", "license_family": "permissive", "maturity": "active", "member_repos": ["bytedance/flux"], "name": "bytedance/flux", "platform": ["python", "cpp"], "pushed_at": "2025-08-28T06:26:38+00:00", "repo": "bytedance/flux", "stars": 1354, "tags": ["communication-overlapping", "tensor-parallelism", "expert-parallelism", "moe", "nvshmem", "cutlass", "pytorch-kernels", "gpu", "cuda", "linux"], "topics": ["cutlass", "pytorch", "cuda", "gpu"], "urls": [], "use_cases": ["overlap allreduce communication with gemm computation during tensor parallel training", "speed up MoE expert parallelism by hiding dispatch/combine communication", "integrate fast communication-overlapping kernels into an existing PyTorch training framework", "reduce GPU idle time in multi-node LLM training", "train dense or mixture-of-experts models more efficiently on Hopper or Ampere GPUs"], "what_it_is": "Flux is a GPU kernel library from ByteDance that overlaps computation with communication for tensor and expert parallelism in dense and MoE model training and inference. It provides high-performance, pluggable CUDA kernels compatible with PyTorch across multiple Nvidia GPU architectures.", "when_to_avoid": ["you are doing single-GPU workloads with no distributed communication to overlap", "you use non-Nvidia GPUs or unsupported CUDA architectures", "you need a full training framework rather than a kernel library"], "when_to_choose": ["you are training or serving large dense or MoE models with tensor/expert parallelism on Nvidia GPUs", "you need pluggable PyTorch-compatible kernels to overlap communication with computation", "you want to squeeze more performance out of multi-GPU or multi-node training"]}, "data_as_of": "2026-08-30T08:39:29.467469+00:00", "members": [{"path": "/products/bytedance-flux", "repo": "bytedance/flux", "role": "main", "score": 33}], "provenance": {"archived": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "artifact_type": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}, "description": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "domain": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}, "enriched": {"inputs": [], "kind": "computed", "method": "enrichment_status"}, "function": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}, "health_score": {"inputs": ["days_since_push", "days_since_release", "archived"], "kind": "computed", "method": "health_v1"}, "homepage": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "language": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "license": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "license_family": {"inputs": ["license"], "kind": "computed", "method": "license_family"}, "maturity": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}, "member_repos": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "name": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "platform": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}, "pushed_at": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "repo": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "stars": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "tags": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}, "topics": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "urls": {"kind": "observed", "observed_at": "2026-08-28T04:04:29.043188+00:00", "source": "github"}, "use_cases": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}, "what_it_is": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}, "when_to_avoid": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}, "when_to_choose": {"confidence": null, "enriched_at": "2026-08-30T04:42:01.176215+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "44bada969f695c3edc8716eb693f463e6ce14d83f0d0e809da94042bdaf732d9", "fetched_at": "2026-08-28T04:04:29.043188+00:00", "kind": "readme", "missing": false, "url": "https://github.com/bytedance/flux"}], "taxonomy_version": 1}}, "score": {"components": {"activity": 39, "longevity": 65, "rhythm": 8}, "computed_at": "2026-09-03T02:39:23.370411+00:00", "flags": [], "formula": "round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)", "inputs": {"age_days": 915, "days_push": 370, "days_rel": 673, "gap_med": null, "n_releases_24m": 1}, "score": 33, "version": 2}, "staleness": {"enrichment_outdated": false, "low_confidence": false, "scrape_days": 9, "stale_scrape": false}}