Commit History
2 days ago
-
Regenerate Diffusers examples with Edge FP8 (#8)
- Regenerate Diffusers examples with Edge FP8 (02aab58fc5176f10dbb26cc7a380405708bcd73b) Co-authored-by: Yiming Zhao <ConstBob@users.noreply.huggingface.co>8ed14ec -
Update vLLM/vLLM-Omni examples; refresh FP8 assets (#7)
- update vllm/vllm-omni example (3dd52e74c641ced5447fddd5e2d32fc26e69ce87) - remove unlinked assets (026da21dea0429afe1e39646ca352176ba88edaa) Co-authored-by: Mu-Ti Chung <mutic@users.noreply.huggingface.co>8669f55 -
Sync Quality Benchmark from reviewed MR !13060
Verified signatureSync only the complete Quality Benchmark section from GitLab MR !13060 at eddc7e7a1f221b6f31ec6c107e32e91d8fc31bea. Preserve all other README sections and repository files unchanged.be2ea4c
3 days ago
-
Integrate reviewed Cosmos3-Edge-FP8 model card with current usage and performance
Verified signatureUse the reviewed MR overview, quality benchmarks, and training/evaluation datasets; retain current HF usage examples and performance measurements. Add the PNG relationship diagram, set release date to November 9, 2026, and correct precision and section links. Exclude internal-only sections. Existing shared subcards are unchanged.542b2c8 -
docs: enable ModelOpt FP8 for Diffusers (#3)
- docs: enable ModelOpt FP8 for Diffusers (b3af554762f228c412a93530806f8cadb9b56917) Co-authored-by: Yiming Zhao <ConstBob@users.noreply.huggingface.co>5ee4fdd -
Add compact links to both Cosmos3 collections
Verified signature5f1f58b
6 days ago
-
Update FP8 performance benchmarks and remove unmeasured T2000 row (#6)
- Update FP8 performance benchmarks and remove unmeasured T2000 row (00213e1b98514e65c963df7fc4806eb72c278ad8) - Remove B200 from model card test hardware list (c9370cc4e13a4017f18da94603b137f4b7b6811e) Co-authored-by: Sergiy Fefilatyev <fsergiy@users.noreply.huggingface.co>02df5ef
24 days ago
-
Consolidate FP8 performance benchmarks by hardware (#2)
- Consolidate FP8 performance benchmarks by hardware (869bc1db266d60fdcbb1e34e961ef698e755b69e) - Use fastest measured image-to-video backend per GPU (768216b3bd2bb0b24539fd4efe7d08702a54ae6e) - Add DROID policy throughput for documented 16-action workload (05b6d7e41fd7c97fe85c4f3e893820d9e7b4c20e) Co-authored-by: Sergiy Fefilatyev <fsergiy@users.noreply.huggingface.co>ed74e10
-
Fix model card image and video links to original model assets
Verified signature8ed186d -
Add FP8 model card based on nvidia/Cosmos3-Edge
Verified signature15a6838