Title: GitHub - ModelTC/SageAttention3-sparse: [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.
Open Graph Title: GitHub - ModelTC/SageAttention3-sparse: [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.
X Title: GitHub - ModelTC/SageAttention3-sparse: [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.
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. - ModelTC/SageAttention3-sparse
Open Graph 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. - Mo...
X 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. - Mo...
Opengraph URL: https://github.com/ModelTC/SageAttention3-sparse
X: @github
Domain: patch-diff.githubusercontent.com
| route-pattern | /:user_id/:repository |
| route-controller | files |
| route-action | disambiguate |
| fetch-nonce | v2:6b74ae2b-2e07-4c25-2c30-dd7a0af69796 |
| current-catalog-service-hash | f3abb0cc802f3d7b95fc8762b94bdcb13bf39634c40c357301c4aa1d67a256fb |
| request-id | B976:85D7F:11CA8A:16F669:698D0567 |
| html-safe-nonce | 7f464e61cf0b9ed2480b346b0c5d13db41d4e2dc26a3b2e8d224a6e5f7e69d74 |
| visitor-payload | eyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiJCOTc2Ojg1RDdGOjExQ0E4QToxNkY2Njk6Njk4RDA1NjciLCJ2aXNpdG9yX2lkIjoiNTYzMDY4NDk2Mjg5NTMxMjM5IiwicmVnaW9uX2VkZ2UiOiJpYWQiLCJyZWdpb25fcmVuZGVyIjoiaWFkIn0= |
| visitor-hmac | 3260d6f7823e3994f2b0b0e508b5cbd901d5df64f4615ca64c111dd473c1230e |
| hovercard-subject-tag | repository:1155149397 |
| github-keyboard-shortcuts | repository,copilot |
| google-site-verification | Apib7-x98H0j5cPqHWwSMm6dNU4GmODRoqxLiDzdx9I |
| octolytics-url | https://collector.github.com/github/collect |
| analytics-location | / |
| fb:app_id | 1401488693436528 |
| apple-itunes-app | app-id=1477376905, app-argument=https://github.com/ModelTC/SageAttention3-sparse |
| twitter:image | https://opengraph.githubassets.com/30a4d6a987c898a8577375f8ad14b514ae00ccef3cc6117ebe8c8abd8ea074a6/ModelTC/SageAttention3-sparse |
| twitter:card | summary_large_image |
| og:image | https://opengraph.githubassets.com/30a4d6a987c898a8577375f8ad14b514ae00ccef3cc6117ebe8c8abd8ea074a6/ModelTC/SageAttention3-sparse |
| og:image:alt | [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. - Mo... |
| og:image:width | 1200 |
| og:image:height | 600 |
| og:site_name | GitHub |
| og:type | object |
| hostname | github.com |
| expected-hostname | github.com |
| None | f2da95634bce8a94cfa4123788169bfabdf845fd1d790fbaaaaab09dcfebdf28 |
| turbo-cache-control | no-preview |
| go-import | github.com/ModelTC/SageAttention3-sparse git https://github.com/ModelTC/SageAttention3-sparse.git |
| octolytics-dimension-user_id | 69665675 |
| octolytics-dimension-user_login | ModelTC |
| octolytics-dimension-repository_id | 1155149397 |
| octolytics-dimension-repository_nwo | ModelTC/SageAttention3-sparse |
| octolytics-dimension-repository_public | true |
| octolytics-dimension-repository_is_fork | true |
| octolytics-dimension-repository_parent_id | 867007699 |
| octolytics-dimension-repository_parent_nwo | thu-ml/SageAttention |
| octolytics-dimension-repository_network_root_id | 867007699 |
| octolytics-dimension-repository_network_root_nwo | thu-ml/SageAttention |
| turbo-body-classes | logged-out env-production page-responsive |
| disable-turbo | false |
| browser-stats-url | https://api.github.com/_private/browser/stats |
| browser-errors-url | https://api.github.com/_private/browser/errors |
| release | c21843b18feba17d11efb1895a7db61e8672f2cf |
| ui-target | full |
| theme-color | #1e2327 |
| color-scheme | light dark |
Links:
Viewport: width=device-width
