add gptq_marlin test for bug report https://github.com/vllm-project/vllm/issues/5088 (#5145)

d919ecc7 · Alexander Matveev · GitHub · e691918e · d919ecc7
Unverified Commit d919ecc7 authored Jun 15, 2024 by Alexander Matveev Committed by GitHub Jun 15, 2024
Hide whitespace changes
Inline Side-by-side

Showing with 3 additions and 0 deletions

tests/models/test_gptq_marlin.py tests/models/test_gptq_marlin.py +3 -0

No files found.
--- a/tests/models/test_gptq_marlin.py
+++ b/tests/models/test_gptq_marlin.py
@@ -40,6 +40,9 @@ MODELS = [
    ("TheBloke/TinyLlama-1.1B-Chat-v1.0-GPTQ", "gptq-8bit-128g-actorder_True"),
    # 8-bit, act_order==True, group_size=32
    ("TheBloke/TinyLlama-1.1B-Chat-v1.0-GPTQ", "gptq-8bit-32g-actorder_True"),
+
+    # 4-bit, act_order==True, group_size=128
+    ("TechxGenus/gemma-1.1-2b-it-GPTQ", "main")
 ]