add quantization tests

vllm-project · Jul 18, 2024 · 7fcac78 · 7fcac78
1 parent fc871a5
commit 7fcac78
Showing 1 changed file with 2 additions and 0 deletions.
diff --git a/tests/basic_correctness/test_cpu_offload.py b/tests/basic_correctness/test_cpu_offload.py
@@ -4,3 +4,5 @@
 def test_cpu_offload():
     compare_two_settings("meta-llama/Llama-2-7b-hf", [],
                          ["--cpu-offload-gb", "4"])
+    compare_two_settings("nm-testing/llama7b-one-shot-2_4-w4a16-marlin24-t",
+                         [], ["--cpu-offload-gb", "1"])