Skip to content

Actions: mudler/vllm.cpp

Actions

containers

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
containers.yml
810 workflow runs
810 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
fix(SPEC-MTP): --speculative-config DROPPED every key it did not read, so probabilistic drafting ran GREEDY in silence (#1160) containers #812: Pull request #1175 synchronize by localai-bot
13m 44s row/FIX-SPEC-CONFIG-UNKNOWN-KEYS
row/FIX-SPEC-CONFIG-UNKNOWN-KEYS
13m 44s
feat(ENG-CUDAGRAPH-DEDUP): fold captures that share a topology onto one graph executable containers #811: Pull request #1178 opened by localai-bot
10m 21s row/ENG-CUDAGRAPH-DEDUP
row/ENG-CUDAGRAPH-DEDUP
10m 21s
spec(ENG-CUDAGRAPH-BREAK): the capture seam has no break points, and the nine hand-rolled drivers each re-derive one containers #810: Pull request #1177 opened by localai-bot
10m 36s row/ENG-CUDAGRAPH-BREAK
row/ENG-CUDAGRAPH-BREAK
10m 36s
fix(SPEC-MTP): --speculative-config DROPPED every key it did not read, so probabilistic drafting ran GREEDY in silence (#1160) containers #809: Pull request #1175 synchronize by localai-bot
4m 58s row/FIX-SPEC-CONFIG-UNKNOWN-KEYS
row/FIX-SPEC-CONFIG-UNKNOWN-KEYS
4m 58s
fix(FIX-GPU-MEM-UTIL-INERT): --gpu-memory-utilization stops reporting success for a budget it discards (#1165) containers #808: Pull request #1176 opened by localai-bot
10m 16s row/FIX-GPU-MEM-UTIL-INERT
row/FIX-GPU-MEM-UTIL-INERT
10m 16s
fix(FIX-FP8-BLOCKWISE-REFUSAL): refuse block-wise FP8 by name at load (#1166) containers #807: Pull request #1172 synchronize by localai-bot
2m 16s row/FIX-FP8-BLOCKWISE-REFUSAL
row/FIX-FP8-BLOCKWISE-REFUSAL
2m 16s
spec(GDN-REPLAYSSM): GDN decode rewrites the whole fp32 state every step, and a replay ring makes the per-step traffic read-only containers #806: Pull request #1174 opened by localai-bot
4m 50s row/SPEC-GDN-REPLAYSSM
row/SPEC-GDN-REPLAYSSM
4m 50s
spec(GDN-MOE-BF16-OUT): the GDN recurrence output and the z gate are f32 on every MoE checkpoint, and both references keep them bf16 containers #805: Pull request #1173 opened by localai-bot
10m 22s row/SPEC-GDN-MOE-BF16-OUT
row/SPEC-GDN-MOE-BF16-OUT
10m 22s
fix(FIX-FP8-BLOCKWISE-REFUSAL): refuse block-wise FP8 by name at load (#1166) containers #804: Pull request #1172 opened by localai-bot
9m 57s row/FIX-FP8-BLOCKWISE-REFUSAL
row/FIX-FP8-BLOCKWISE-REFUSAL
9m 57s
record(ENG-CUDAGRAPH): SGLang's Breakable CUDA Graph, grounded at its pin, and the prefill lever it does not give us (#1161) containers #803: Pull request #1167 opened by localai-bot
22m 43s row/ENG-CUDAGRAPH-BCG-FOLD
row/ENG-CUDAGRAPH-BCG-FOLD
22m 43s
record(GEMMA4-ROCM-KEEP): the 2x R9700 KEEP recipe, and the denominator it still lacks containers #802: Pull request #676 synchronize by localai-bot
35m 55s bakon11:row/GEMMA4-ROCM-KEEP-MEASURED
bakon11:row/GEMMA4-ROCM-KEEP-MEASURED
35m 55s
feat(BACKEND-ROCM): select the attention backend in the runner containers #801: Pull request #1065 synchronize by localai-bot
Action required tbrasser:row/BACKEND-ROCM-ATTN-RUNNER
tbrasser:row/BACKEND-ROCM-ATTN-RUNNER
Action required
feat(BACKEND-ROCM): register a ROCm attention backend for kROCM containers #800: Pull request #1056 synchronize by localai-bot
Action required tbrasser:row/BACKEND-ROCM-ATTN-REGISTER
tbrasser:row/BACKEND-ROCM-ATTN-REGISTER
Action required
fix(rocm/gemma4): #839 retirement-safe prefill peer GeGLU Launch/Finish containers #799: Pull request #1047 synchronize by localai-bot
33m 44s bakon11:fix/gemma4-839-prefill-peer
bakon11:fix/gemma4-839-prefill-peer
33m 44s
fix(ENG-RELEASE-WINDOWS): the api-server gate can report its own failure again containers #798: Pull request #1159 opened by localai-bot
30m 18s row/ENG-RELEASE-WINDOWS-test-thread-raii
row/ENG-RELEASE-WINDOWS-test-thread-raii
30m 18s
measure(#810 A3): the token gate ran on real weights and FAILED 6/96, and the failure names the mechanism containers #797: Pull request #1158 opened by localai-bot
59m 21s row/MODEL-NEMOTRON-H-A3-RESULT
row/MODEL-NEMOTRON-H-A3-RESULT
59m 21s
feat(LTX25-TI2VID-RECIPE): the plain two-stage pipeline, on the schedule anchor upstream actually uses (#1093) containers #796: Pull request #1156 synchronize by localai-bot
15m 9s row/LTX25-TI2VID-RECIPE
row/LTX25-TI2VID-RECIPE
15m 9s
feat(BACKEND-TENSTORRENT-HOST-FREE-FORWARD): capture the Qwen3 decode graph and advance cur_pos on device (#1105) containers #795: Pull request #1053 synchronize by localai-bot
24m 22s lu-zero:row/BACKEND-TENSTORRENT-HOST-FREE-R2
lu-zero:row/BACKEND-TENSTORRENT-HOST-FREE-R2
24m 22s
build: make the tree compile on gcc 16, and add a CI lane so it stays that way containers #794: Pull request #945 synchronize by localai-bot
28m 12s jimmykarily:row/BUILD-GCC16
jimmykarily:row/BUILD-GCC16
28m 12s
feat(LTX25-TI2VID-RECIPE): the plain two-stage pipeline, on the schedule anchor upstream actually uses (#1093) containers #793: Pull request #1156 synchronize by localai-bot
18m 46s row/LTX25-TI2VID-RECIPE
row/LTX25-TI2VID-RECIPE
18m 46s
build: make the tree compile on gcc 16, and add a CI lane so it stays that way containers #792: Pull request #945 synchronize by localai-bot
Action required jimmykarily:row/BUILD-GCC16
jimmykarily:row/BUILD-GCC16
Action required
feat(LTX25-TI2VID-RECIPE): the plain two-stage pipeline, on the schedule anchor upstream actually uses (#1093) containers #791: Pull request #1156 opened by localai-bot
51m 21s row/LTX25-TI2VID-RECIPE
row/LTX25-TI2VID-RECIPE
51m 21s
fix(ENG-RELEASE-WINDOWS): the main baseline runs the two MSVC gates it grades (#503) containers #790: Pull request #1155 synchronize by localai-bot
49m 51s row/ENG-RELEASE-WINDOWS-baseline-coverage
row/ENG-RELEASE-WINDOWS-baseline-coverage
49m 51s
feat(ENG-RESIDENCY-CONFIG): the disk-residency tier is a config key, an absent field means unchanged at BOTH ends of the install, and a typo anywhere in the document is refused (#1110, #1109, #1122, #1133) containers #788: Pull request #1119 synchronize by localai-bot
43m 6s row/ENG-RESIDENCY-CONFIG
row/ENG-RESIDENCY-CONFIG
43m 6s
feat(BACKEND-TENSTORRENT-HOST-FREE-FORWARD): capture the Qwen3 decode graph and advance cur_pos on device (#1105) containers #787: Pull request #1053 synchronize by localai-bot
35m 11s lu-zero:row/BACKEND-TENSTORRENT-HOST-FREE-R2
lu-zero:row/BACKEND-TENSTORRENT-HOST-FREE-R2
35m 11s