A high-throughput and memory-efficient inference and serving engine for LLMs
[CI][NIXL] Pin NIXL to 1.2.0 (#45843)
Signed-off-by: Itay Alroy <ialroy@nvidia.com> Signed-off-by: Itay Alroy <75032521+itayalroy@users.noreply.github.com> Co-authored-by: ovidiusm <ovidium@nvidia.com>
Itay Alroy committed
Jun 17, 2026 at 07:29 UTC
d78650cf973b9a1d8c03160c691f98725f8edb16
1 file changed
+1
-1
requirements/kv_connectors.txt
+1
-1
@@ -2,5 +2,5 @@ lmcache >= 0.3.9
2
# CuPy 14.1.0 imports pytest from cupy.testing._random. Use <14.1.0
3
# until a fixed newer release is verified for runtime images.
4
cupy-cuda13x < 14.1.0
5
-nixl >= 1.1.0 # Required for disaggregated prefill
5
+nixl == 1.2.0 # Required for disaggregated prefill
6
mooncake-transfer-engine >= 0.3.8