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 17, 05:30 UTC
[Minor] Fix test_cache.py CI test failure (#2684)
Philipp Moritz committed
Jan 31, 2024 at 10:12 UTC
89efcf1ce53cd01c27384e3c3e1c6b0761978076
1 file changed
+2
-2
tests/kernels/test_cache.py
+2
-2
@@ -200,12 +200,12 @@ def test_swap_blocks(
200
201
# Create the KV caches on the first device.
202
src_key_caches, src_value_caches = kv_cache_factory(
203
- num_blocks, block_size, 1, num_heads, head_size, dtype, seed,
203
+ num_blocks, block_size, 1, num_heads, head_size, dtype, None, seed,
204
src_device)
205
206
# Create the KV caches on the second device.
207
dist_key_caches, dist_value_caches = kv_cache_factory(
208
- num_blocks, block_size, 1, num_heads, head_size, dtype, seed,
208
+ num_blocks, block_size, 1, num_heads, head_size, dtype, None, seed,
209
dst_device)
210
211
src_key_caches_clone = src_key_caches[0].clone()