Skip to content

Actions: SJTU-DENG-Lab/Diffulex

Actions

Documentation

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
publish-docs.yml
63 workflow runs
63 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
docs: update README to include DMax-specific JSON task files Documentation #63: Commit bdeaa28 pushed by drewjin
21s main
main
21s
pref: make code eval efficient Documentation #62: Commit f29dc43 pushed by drewjin
30s main
main
30s
refactor: update max_nfe handling and enhance logging in MultiBlockSc… Documentation #61: Commit 1041a3d pushed by drewjin
25s main
main
25s
fix: make sure arg confirm-run-unsafe-code passed into bench args Documentation #60: Commit 3ba3a03 pushed by drewjin
29s main
main
29s
feat: refactor attention implementation and enhance model runner capa… Documentation #59: Commit d0c83e1 pushed by drewjin
22s main
main
22s
chore: update configs and scripts Documentation #58: Commit 0a0ba03 pushed by drewjin
40s main
main
40s
feat: enhance configuration handling and improve request processing Documentation #57: Commit e4eb9b9 pushed by drewjin
21s main
main
21s
feat: add full static runner support and enhance model execution Documentation #56: Commit d3d8d08 pushed by drewjin
28s main
main
28s
Revert "feat: add support for inputs_embeds in CUDA graph execution" Documentation #55: Commit a1a089e pushed by drewjin
41s main
main
41s
feat: add support for inputs_embeds in CUDA graph execution Documentation #54: Commit bdac9ed pushed by drewjin
26s main
main
26s
chore: update config Documentation #53: Commit 22a626f pushed by drewjin
21s main
main
21s
feat: add full static runner support and enhance model execution Documentation #52: Commit d3d8d08 pushed by drewjin
25s main
main
25s
feat: enhance model configuration and input handling Documentation #51: Commit d9d3eb9 pushed by drewjin
29s main
main
29s
feat: add full static runner support and enhance model execution Documentation #50: Commit d3d8d08 pushed by drewjin
21s main
main
21s
feat: introduce vLLM compatibility layer for improved model execution Documentation #49: Commit 782148d pushed by drewjin
29s main
main
29s
feat: implement workspace caching for fused MoE in vllm_fuse_moe.py Documentation #48: Commit 313883e pushed by drewjin
26s main
main
26s
fix: adjust intermediate cache size calculation in fused_experts_impl Documentation #47: Commit 8a60621 pushed by drewjin
26s main
main
26s
chore: modify configs Documentation #46: Commit fa018f3 pushed by drewjin
27s main
main
27s
feat: update README and remove deprecated harness scripts Documentation #45: Commit 208d144 pushed by drewjin
24s main
main
24s
chore: update gitignore Documentation #44: Commit fa675c8 pushed by drewjin
29s main
main
29s
feat: add DMax-specific utilities and configurations for HumanEval, M… Documentation #43: Commit d34dbb8 pushed by drewjin
42s main
main
42s
feat: add new submodules and enhance project configuration Documentation #42: Commit cafe9b7 pushed by drewjin
31s main
main
31s
feat: enhance LLaDA2 model with mask token handling and CUDA graph su… Documentation #41: Commit 0b9c6a1 pushed by drewjin
24s main
main
24s
fix: avoid repeated req initialization Documentation #40: Commit cdcf964 pushed by drewjin
29s main
main
29s
feat: implement editable start functionality in DllmBlock Documentation #39: Commit 51baf06 pushed by drewjin
42s main
main
42s