{"adoption": {"forks": 491, "observed_at": "2026-08-28T04:08:13.822599+00:00", "stars": 3684}, "canonical_url": "https://ross.abutalabs.com/products/sageattention", "card": {"archived": false, "artifact_type": "library", "description": "[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.", "domain": ["large-language-models", "machine-learning", "gpu-computing", "image-processing"], "enriched": true, "function": ["llm-inference", "machine-learning", "gpu-computing", "deep-learning"], "health_score": 69, "homepage": "https://arxiv.org/abs/2410.02367", "language": "Cuda", "license": "Apache-2.0", "license_family": "permissive", "maturity": "active", "member_repos": ["thu-ml/SageAttention"], "name": "thu-ml/SageAttention", "platform": ["python", "windows"], "pushed_at": "2026-01-17T17:43:28+00:00", "repo": "thu-ml/SageAttention", "stars": 3684, "tags": ["quantization", "attention-kernels", "flashattention", "inference-acceleration", "int8", "fp8", "fp4", "triton", "video-generation", "diffusion-models", "plug-and-play", "video", "gpu", "cuda", "linux"], "topics": ["attention", "inference-acceleration", "llm", "quantization", "cuda", "triton", "video-generation", "efficient-attention", "mlsys", "llm-infra", "vit", "video-generate"], "urls": [], "use_cases": ["speed up LLM inference attention computation", "accelerate video generation model inference", "quantize attention to int8 or fp8 for faster transformers", "replace flashattention with a faster drop-in kernel", "reduce inference latency for diffusion transformer models", "run ViT models faster on consumer GPUs"], "what_it_is": "SageAttention is a family of quantized attention kernels (INT8/FP8/FP4) that accelerate transformer inference 2-5x over FlashAttention with negligible accuracy loss. It works plug-and-play across language, image, and video generation models on Ampere, Ada, Hopper, and Blackwell GPUs.", "when_to_avoid": ["you need bit-exact attention outputs for precision-sensitive training or evaluation", "you target non-NVIDIA hardware like AMD or Apple GPUs", "your model uses attention variants unsupported by the kernels"], "when_to_choose": ["you need faster transformer inference on NVIDIA GPUs without retraining", "you run LLM, image, or video generation models and attention is the bottleneck", "you want plug-and-play acceleration compatible with torch.compile and distributed inference"]}, "data_as_of": "2026-08-30T08:39:29.467469+00:00", "members": [{"path": "/products/sageattention", "repo": "thu-ml/SageAttention", "role": "main", "score": 40}], "provenance": {"archived": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "artifact_type": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}, "description": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "domain": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}, "enriched": {"inputs": [], "kind": "computed", "method": "enrichment_status"}, "function": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}, "health_score": {"inputs": ["days_since_push", "days_since_release", "archived"], "kind": "computed", "method": "health_v1"}, "homepage": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "language": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "license": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "license_family": {"inputs": ["license"], "kind": "computed", "method": "license_family"}, "maturity": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}, "member_repos": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "name": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "platform": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}, "pushed_at": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "repo": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "stars": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "tags": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}, "topics": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "urls": {"kind": "observed", "observed_at": "2026-08-28T04:08:13.822599+00:00", "source": "github"}, "use_cases": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}, "what_it_is": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}, "when_to_avoid": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}, "when_to_choose": {"confidence": null, "enriched_at": "2026-08-29T18:31:11.557830+00:00", "kind": "inferred", "prompt_version": 1, "sources": [{"content_hash": "aa6226e339e1b67003b05986e44fce9f4e68b5770a85c8298efe068b07858e8e", "fetched_at": "2026-08-28T04:08:13.822599+00:00", "kind": "readme", "missing": false, "url": "https://github.com/thu-ml/SageAttention"}, {"content_hash": "da51c7a1a436a693fb014a2f4209789d6919e7887910b76f19e03aa9a5bd8afe", "fetched_at": "2026-08-29T09:25:01.917440+00:00", "kind": "homepage", "missing": false, "url": "https://arxiv.org/abs/2410.02367"}, {"content_hash": "cca9c3a11c56b1212864e03fb8a5511a3bdb6df53398ba4dd2ed778f931607dd", "fetched_at": "2026-08-29T09:25:01.919885+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/donate.html"}, {"content_hash": "47cbc55ff1deaca3f36b463373de863a7053d91cacf9f95725618e11610209ae", "fetched_at": "2026-08-29T09:25:01.922988+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about/ourmembers.html"}, {"content_hash": "a1f16f915a9ad89ee57c3e929a1b6577d481f52c05f8015fc720b16ee8014f19", "fetched_at": "2026-08-29T09:25:01.925053+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/about"}, {"content_hash": "b14a8d05a0ec337409ab63bffd91306b5317b16fac358aa6d1fe6afbaf3c8628", "fetched_at": "2026-08-29T09:25:01.921453+00:00", "kind": "site_page", "missing": false, "url": "https://info.arxiv.org/labs/index.html"}], "taxonomy_version": 1}}, "score": {"components": {"activity": 62, "longevity": 49, "rhythm": 8}, "computed_at": "2026-09-02T17:46:02.011165+00:00", "flags": ["prerelease_only"], "formula": "round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)", "inputs": {"age_days": 699, "days_push": 228, "days_rel": 582, "gap_med": null, "n_releases_24m": 1}, "score": 40, "version": 2}, "staleness": {"enrichment_outdated": false, "low_confidence": false, "scrape_days": 9, "stale_scrape": false}}