Skip to content

Actions: xLLM-AI/xllm

Actions

xLLM Build arm_64 NPU

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
build_arm_64_npu.yaml
2,500+ workflow runs
2,500+ 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
feat: add xlite as third npu backend (support qwen3-dense/qwen3-moe/glm-4.7/glm-5.1/glm-5.2) xLLM Build arm_64 NPU #3139: Pull request #2282 submitted by yinjiawei01
Queued
Queued
feat: add xlite as third npu backend (support qwen3-dense/qwen3-moe/glm-4.7/glm-5.1/glm-5.2) xLLM Build arm_64 NPU #3138: Pull request #2282 submitted by yt-zeng
3s
3s
feat: add xlite as third npu backend (support qwen3-dense/qwen3-moe/glm-4.7/glm-5.1/glm-5.2) xLLM Build arm_64 NPU #3137: Pull request #2282 submitted by ustcfy
4s
4s
feat: add xlite as third npu backend (support qwen3-dense/qwen3-moe/glm-4.7/glm-5.1/glm-5.2) xLLM Build arm_64 NPU #3136: Pull request #2282 submitted by liutongxuan
4h 55m 15s
4h 55m 15s
feat: support qwen3 dspark decoding on NPU with Python executor. xLLM Build arm_64 NPU #3135: Pull request #2263 submitted by zhang-minchao
1d 0h 0m 3s
1d 0h 0m 3s
feat: support qwen3 dspark decoding on NPU with Python executor. xLLM Build arm_64 NPU #3134: Pull request #2263 submitted by ustcfy
44m 5s
44m 5s
feat: support Qwen3.5 Decode Context Parallel on Ascend NPU. xLLM Build arm_64 NPU #3133: Pull request #2125 synchronize by Enguikong
1d 0h 0m 3s Enguikong:feat/qwen35-dcp-correctness
Enguikong:feat/qwen35-dcp-correctness
1d 0h 0m 3s
feat: support qwen3 dspark decoding on NPU with Python executor. xLLM Build arm_64 NPU #3132: Pull request #2263 submitted by weizhehuang0827
56m 54s
56m 54s
feat: add lag confidence to decouple adaptive speculative pruning from critical path. xLLM Build arm_64 NPU #3131: Pull request #2280 synchronize by weizhehuang0827
1d 0h 0m 2s weizhehuang0827:feat/adaptive-lag-confidence
weizhehuang0827:feat/adaptive-lag-confidence
1d 0h 0m 2s
feat: support Qwen3.5 Decode Context Parallel on Ascend NPU. xLLM Build arm_64 NPU #3130: Pull request #2125 synchronize by Enguikong
55m 59s Enguikong:feat/qwen35-dcp-correctness
Enguikong:feat/qwen35-dcp-correctness
55m 59s
feat: support qwen3 dspark decoding on NPU with Python executor. xLLM Build arm_64 NPU #3129: Pull request #2263 synchronize by yinjiawei01
1h 40m 34s yinjiawei01:features/qwen3_dspark_python_npu
yinjiawei01:features/qwen3_dspark_python_npu
1h 40m 34s
feat: support Qwen3.5 Decode Context Parallel on Ascend NPU. xLLM Build arm_64 NPU #3128: Pull request #2125 synchronize by Enguikong
1h 48m 34s Enguikong:feat/qwen35-dcp-correctness
Enguikong:feat/qwen35-dcp-correctness
1h 48m 34s
feat: support Qwen3.5 Decode Context Parallel on Ascend NPU. xLLM Build arm_64 NPU #3127: Pull request #2125 synchronize by Enguikong
2h 40m 20s Enguikong:feat/qwen35-dcp-correctness
Enguikong:feat/qwen35-dcp-correctness
2h 40m 20s
bugfix: separate greedy and probabilistic draft proposals for rejecti… xLLM Build arm_64 NPU #3126: Commit 9e0f140 pushed by zhang-minchao
1d 0h 0m 2s main
main
1d 0h 0m 2s
feat: append vit executor for qwen3 vl model xLLM Build arm_64 NPU #3125: Pull request #2207 synchronize by AmpereKowk
Action required AmpereKowk:qwen3_vl_executor
AmpereKowk:qwen3_vl_executor
Action required
feat: append vit executor for qwen3 vl model xLLM Build arm_64 NPU #3124: Pull request #2207 synchronize by AmpereKowk
Action required AmpereKowk:qwen3_vl_executor
AmpereKowk:qwen3_vl_executor
Action required
feat: support EPLB for DeepSeek-V3.2 Python model executor. xLLM Build arm_64 NPU #3123: Pull request #2283 opened by yingxudeng
1d 0h 0m 1s yingxudeng:feat/deepseek-v32-python-eplb
yingxudeng:feat/deepseek-v32-python-eplb
1d 0h 0m 1s
bugfix: separate greedy and probabilistic draft proposals for rejection sampling. xLLM Build arm_64 NPU #3122: Pull request #2175 submitted by zhang-minchao
1d 0h 0m 3s
1d 0h 0m 3s
bugfix: separate greedy and probabilistic draft proposals for rejection sampling. xLLM Build arm_64 NPU #3121: Pull request #2175 submitted by DongheJin
44m 30s
44m 30s
feat: add opt-in FIA decode for Qwen3.5. xLLM Build arm_64 NPU #3120: Pull request #2240 submitted by Enguikong
1d 0h 0m 3s
1d 0h 0m 3s
feat: add opt-in FIA decode for Qwen3.5. xLLM Build arm_64 NPU #3119: Pull request #2240 synchronize by Enguikong
7m 53s Enguikong:feat/qwen35-fia-decode
Enguikong:feat/qwen35-fia-decode
7m 53s
bugfix: make DSV4 SWA capacity PD-aware. xLLM Build arm_64 NPU #3118: Pull request #2265 synchronize by Kang-Meng
1d 0h 0m 2s Kang-Meng:bugfix_swa
Kang-Meng:bugfix_swa
1d 0h 0m 2s
bugfix: make DSV4 SWA capacity PD-aware. xLLM Build arm_64 NPU #3117: Pull request #2265 synchronize by Kang-Meng
5m 25s Kang-Meng:bugfix_swa
Kang-Meng:bugfix_swa
5m 25s
bugfix: make DSV4 SWA capacity PD-aware. xLLM Build arm_64 NPU #3116: Pull request #2265 synchronize by Kang-Meng
24m 46s Kang-Meng:bugfix_swa
Kang-Meng:bugfix_swa
24m 46s
feat: add xlite as third npu backend (support qwen3-dense/qwen3-moe/glm-4.7/glm-5.1/glm-5.2) xLLM Build arm_64 NPU #3115: Pull request #2282 synchronize by banxiaduhuo
4s banxiaduhuo:main
banxiaduhuo:main
4s