A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 5, 03:30 UTC
[CI][NIXL] Fix PD CI breakage: pin nixl-cu{12,13} versions (#39851)
Signed-off-by: ZhanqiuHu <zhu@redhat.com>
zhanqiuhu committed
Apr 15, 2026 at 02:50 UTC
799973af4e618190bfeae053517c0f53c9f8be96
1 file changed
+2
requirements/kv_connectors.txt
+2
@@ -1,3 +1,5 @@
1
lmcache >= 0.3.9
2
nixl[cu13] >= 0.7.1, < 0.10.0 # Required for disaggregated prefill
3
+nixl-cu12 >= 0.7.1, < 0.10.0
4
+nixl-cu13 >= 0.7.1, < 0.10.0
5
mooncake-transfer-engine >= 0.3.8