This website requires JavaScript.
Explore
Help
Register
Sign In
423A35C7
0 Followers
·
0 Following
Joined on
2024-05-11
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
User to block:
Optional note:
The note is not visible to the blocked user.
Cancel
Block
Repositories
8
Projects
Packages
Public Activity
Starred Repositories
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-10-10 16:55:45 +08:00
1fded918ac
[docs] add security notes to NPU Docker overviews (
#10883
)
423A35C7
synced commits to
main
at
423A35C7/pytorch3d
from mirror
2026-09-30 19:25:46 +08:00
88e182f989
Suppress type errors for Pyre upgrade
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-09-28 18:45:46 +08:00
ce9dc9e072
[docker] add Ascend 950PR&950DT images (
#10851
)
c1dc0f51ab
[model] add GLM-5.3-Flash training support (
#10840
)
4d6c7cf03b
Intel docker file (
#10753
)
d4823eb10e
[xpu] add Intel XPU support to test infrastructure and runs_on markers (
#10835
)
Compare 4 commits »
423A35C7
synced commits to
main
at
423A35C7/pytorch3d
from mirror
2026-09-15 19:55:47 +08:00
978cd99221
replace IASF/FASI pulsar macros with inline bit-cast functions
716c594eab
Fix symeig3x3 for inputs that are not of order 1 (
#2045
)
87a6e51e34
Speed up save_obj text formatting, tidy _save helpers
fa22ad16c4
Allow saving OBJ UVs without a texture map (
#2047
)
Compare 4 commits »
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-09-14 19:25:43 +08:00
97b32d3133
[v1] update docs (
#10684
)
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-09-10 01:05:48 +08:00
100e9a42c6
[KT] Support Kimi K2.5/2.6 LoRA fine-tuning (
#10826
)
31078aa10a
[v1] Support multimodal Ulysses CP and memory-efficient chunk loss for SFT (
#10762
)
Compare 2 commits »
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-09-08 16:25:45 +08:00
673048c6a5
[v1] fix non-persistent buffer sync for init_on_rank0 (
#10820
)
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-09-04 22:35:47 +08:00
dced5f8804
[train] support HyperParallel Expert-Parallel (
#10811
)
423A35C7
synced commits to
main
at
423A35C7/pytorch3d
from mirror
2026-09-03 21:55:44 +08:00
0a7d4c1a17
Check CUDA device calls in Pulsar renderer
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-09-02 21:35:45 +08:00
4451765a6b
[model] add Qwen3.8 model support (
#10749
)
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-08-31 20:35:43 +08:00
d6bb97ddff
[data] fix position ids after merging packed-mrope (
#10783
)
6369f90cee
docs: refine PenguinHarness star invitation (
#10806
)
e4d79d2993
docs: encourage following and starring PenguinHarness (
#10805
)
6f38e73b82
[data] add minicpm5 template with XML tool calling (
#10801
)
Compare 4 commits »
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-08-28 02:45:44 +08:00
7fcf5b3b13
[v1] support LoRA with FSDPTurbo expert parallelism (
#10791
)
423A35C7
synced commits to
main
at
423A35C7/pytorch3d
from mirror
2026-08-28 02:35:43 +08:00
e73a7e7bfc
Fix coordinate indexing in frustum face culling (
#2044
)
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-08-27 10:25:44 +08:00
273a988ebe
docs: refresh PenguinHarness description (
#10793
)
423A35C7
synced commits to
main
at
423A35C7/pytorch3d
from mirror
2026-08-27 02:05:43 +08:00
a081d766b0
Fix flaky ICP test_heterogeneous_inputs
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-08-25 17:35:44 +08:00
a18110d2f0
[docker] add NPU image tag history (
#10786
)
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-08-20 23:15:47 +08:00
c4e09c7cbe
[v1] support GDN Ulysses cp (
#10727
)
423A35C7
synced commits to
main
at
423A35C7/pytorch3d
from mirror
2026-08-19 14:25:44 +08:00
fdaf9bd6fe
Remove unused type error suppressions
423A35C7
synced commits to
main
at
423A35C7/LLaMA-Factory
from mirror
2026-08-18 22:15:45 +08:00
ff6d4d12ee
[train] Add KTransformers VLM fine-tuning support (
#10760
)
423A35C7
synced commits to
main
at
423A35C7/pytorch3d
from mirror
2026-08-18 05:45:44 +08:00
2bce7110d5
Fix and speed up the fast axis-angle conversions
a5093b158b
Fix CPU marching cubes precision and topology at high resolutions (
#1934
) (
#2043
)
Compare 2 commits »
First
Previous
1
2
3
4
5
...
Next
Last
