Default Branch

main 3125d79950 · [Chore] Remove unused PolyNorm layer (#27110) · Updated 2025-10-18 03:03:43 +08:00

ZIP  TAR.GZ

Branches

mla-support-awq-marlin 243408b6b4 · Support moe_wna16 as well · Updated 2025-02-13 03:18:29 +08:00	7967 4		ZIP  TAR.GZ
tpu_v1_optimized 70b4e46e70 · compilation is fixed · Updated 2025-02-07 04:49:29 +08:00	8233 14		ZIP  TAR.GZ
v0.7.2-staging-branch 0408efc6d0 · [Misc] Improve error message for incorrect pynvml (#12809) · Updated 2025-02-06 15:23:50 +08:00	-1 -1		ZIP  TAR.GZ
full_cudagraph 1244c25908 · minimize fill_ · Updated 2025-02-05 06:03:51 +08:00	8115 13		ZIP  TAR.GZ
mla_cuda_graphs 0a02744dc8 · fix TP · Updated 2025-01-31 09:18:56 +08:00    xinyun	8161 12		ZIP  TAR.GZ
qwen25vl 0405645a6c · initial · Updated 2025-01-31 08:55:49 +08:00    xinyun	8159 1		ZIP  TAR.GZ
tpu_v1 39c4a4cdb5 · review comments · Updated 2025-01-29 07:08:50 +08:00    xinyun	8233 7		ZIP  TAR.GZ
moondream2 a7ca0cc47f · Merge branch 'main' into moondream2 · Updated 2025-01-20 16:10:52 +08:00    xinyun	8309 2		ZIP  TAR.GZ
v1-blocktable-opt 7097f31955 · test · Updated 2025-01-15 19:22:32 +08:00    xinyun	-1 -1		ZIP  TAR.GZ
correct-docs-cuda-version c1d1875ba3 · Updates docs with correction about default cuda version · Updated 2025-01-08 06:29:07 +08:00    xinyun	8479 1		ZIP  TAR.GZ
torch_dynamo 617fb893d5 · add compile · Updated 2024-07-27 10:29:36 +08:00    xinyun	10434 1		ZIP  TAR.GZ
optimize-prefix-caching-scheduling d5bf492f16 · Merge branch 'main' into optimize-prefix-caching-scheduling · Updated 2024-06-04 08:20:15 +08:00    xinyun	11056 4		ZIP  TAR.GZ
fix-hashing-partial-blocks 1936d7bab0 · format · Updated 2024-06-02 08:02:54 +08:00    xinyun	11074 2		ZIP  TAR.GZ
jax-tpu c00ddd6834 · Add buffer donation to benchmark · Updated 2024-05-01 05:58:47 +08:00    xinyun	11409 75		ZIP  TAR.GZ