Skip to content

Actions: vllm-project/vllm-gguf-plugin

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
297 workflow runs
297 workflow runs

Filter by Workflow

⌥ + click/return to exclude

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
[Model] Add Qwen-VL GGUF support pre-commit #359: Pull request #140 opened by tangzzycc
Action required tangzzycc:feat-qwen-vl-gguf
tangzzycc:feat-qwen-vl-gguf
Action required
[Model] Support Wan2.2 TI2V-5B GGUF diffusion weights pre-commit #358: Pull request #138 synchronize by 0z5a
Action required 0z5a:codex/wan22-gguf
0z5a:codex/wan22-gguf
Action required
[Bugfix] Preserve non-GGUF diffusion loader keyword arguments pre-commit #357: Pull request #139 synchronize by 0z5a
Action required 0z5a:codex/diffusion-quant-loader-kwargs
0z5a:codex/diffusion-quant-loader-kwargs
Action required
[Model] Support Wan2.2 TI2V-5B GGUF diffusion weights pre-commit #356: Pull request #138 synchronize by 0z5a
Action required 0z5a:codex/wan22-gguf
0z5a:codex/wan22-gguf
Action required
[Bugfix] Preserve non-GGUF diffusion loader keyword arguments pre-commit #355: Pull request #139 synchronize by 0z5a
Action required 0z5a:codex/diffusion-quant-loader-kwargs
0z5a:codex/diffusion-quant-loader-kwargs
Action required
fix(moe_vec): chunk launches when tokens*top_k exceeds gridDim.z limit pre-commit #354: Pull request #125 synchronize by BruceLoveDecimal
Action required BruceLoveDecimal:fix/moe_vec_grid_z
BruceLoveDecimal:fix/moe_vec_grid_z
Action required
fix(moe_vec): chunk launches when tokens*top_k exceeds gridDim.z limit pre-commit #353: Pull request #125 synchronize by BruceLoveDecimal
Action required BruceLoveDecimal:fix/moe_vec_grid_z
BruceLoveDecimal:fix/moe_vec_grid_z
Action required
[BugFix] Correct the iq3xs_grid table for IQ3_S dequantization (#136) pre-commit #352: Commit e2b8ad5 pushed by Isotr0py
44s main
main
44s
[Model] Support Wan2.2 TI2V-5B GGUF diffusion weights pre-commit #351: Pull request #138 synchronize by 0z5a
Action required 0z5a:codex/wan22-gguf
0z5a:codex/wan22-gguf
Action required
[Bugfix] Preserve non-GGUF diffusion loader keyword arguments pre-commit #350: Pull request #139 opened by 0z5a
Action required 0z5a:codex/diffusion-quant-loader-kwargs
0z5a:codex/diffusion-quant-loader-kwargs
Action required
[Model] Support Wan2.2 TI2V-5B GGUF diffusion weights pre-commit #349: Pull request #138 opened by 0z5a
Action required 0z5a:codex/wan22-gguf
0z5a:codex/wan22-gguf
Action required
[BugFix] Correct the iq3xs_grid table for IQ3_S dequantization pre-commit #348: Pull request #136 opened by axiom-of-choice
36s axiom-of-choice:fix-iq3-xs-grid
axiom-of-choice:fix-iq3-xs-grid
36s
[BugFix] Dispatch each GGUF MoE tensor to the best kernel its own quant type supports pre-commit #347: Pull request #135 opened by KaigeGao1110
Action required KaigeGao1110:pr4-per-tensor-moe-dispatch
KaigeGao1110:pr4-per-tensor-moe-dispatch
Action required
[BugFix] Resolve Triton MoE BLOCK_M from the per-type table pre-commit #346: Pull request #134 opened by KaigeGao1110
Action required KaigeGao1110:pr3-triton-moe-block-m
KaigeGao1110:pr3-triton-moe-block-m
Action required
[BugFix] Treat negative expert ids as empty routing slots in GGUF MoE pre-commit #345: Pull request #133 opened by KaigeGao1110
2m 45s KaigeGao1110:pr2-moe-negative-expert-ids
KaigeGao1110:pr2-moe-negative-expert-ids
2m 45s
[BugFix] Split per-row MoE launches at CUDA's grid z limit pre-commit #344: Pull request #132 opened by KaigeGao1110
3m 23s KaigeGao1110:pr1-moe-vec-grid-z
KaigeGao1110:pr1-moe-vec-grid-z
3m 23s
fix(weight_utils): name weight_type companions after the last weight only pre-commit #343: Pull request #131 opened by KaigeGao1110
Action required KaigeGao1110:fix/gguf-weight-type-companion-name
KaigeGao1110:fix/gguf-weight-type-companion-name
Action required
fix(fused_moe): treat negative expert ids as empty routing slots pre-commit #342: Pull request #130 opened by KaigeGao1110
Action required KaigeGao1110:fix/gguf-moe-negative-expert-ids
KaigeGao1110:fix/gguf-moe-negative-expert-ids
Action required
fix(gguf): validate every dim/stride crossing into int kernel params (fixes #127) pre-commit #341: Pull request #128 synchronize by x14ngch3n
Action required x14ngch3n:fix/int32-range-kernel-params
x14ngch3n:fix/int32-range-kernel-params
Action required
fix(gguf): validate every dim/stride crossing into int kernel params (fixes #127) pre-commit #340: Pull request #128 synchronize by x14ngch3n
Action required x14ngch3n:fix/int32-range-kernel-params
x14ngch3n:fix/int32-range-kernel-params
Action required
fix(gguf): validate every dim/stride crossing into int kernel params (fixes #127) pre-commit #339: Pull request #128 synchronize by x14ngch3n
-1s x14ngch3n:fix/int32-range-kernel-params
x14ngch3n:fix/int32-range-kernel-params
-1s
fix(gguf): validate every dim/stride crossing into int kernel params (fixes #127) pre-commit #338: Pull request #128 opened by x14ngch3n
Action required x14ngch3n:fix/int32-range-kernel-params
x14ngch3n:fix/int32-range-kernel-params
Action required
[Models] Support DeepSeek-V4 GGUF weights mapping and architecture fallback pre-commit #337: Pull request #126 opened by DjoserKhemSimeu
Action required DjoserKhemSimeu:feat/deepseek-v4-support
DjoserKhemSimeu:feat/deepseek-v4-support
Action required
fix(moe_vec): chunk launches when tokens*top_k exceeds gridDim.z limit pre-commit #336: Pull request #125 reopened by BruceLoveDecimal
1m 24s BruceLoveDecimal:fix/moe_vec_grid_z
BruceLoveDecimal:fix/moe_vec_grid_z
1m 24s
fix(moe_vec): chunk launches when tokens*top_k exceeds gridDim.z limit pre-commit #335: Pull request #125 opened by BruceLoveDecimal
2m 18s BruceLoveDecimal:fix/moe_vec_grid_z
BruceLoveDecimal:fix/moe_vec_grid_z
2m 18s