From 01c16ede6b4fc2e07b6eb5a4f60a64a1f365e460 Mon Sep 17 00:00:00 2001 From: Michael Goin Date: Tue, 23 Jul 2024 18:45:12 -0400 Subject: [PATCH] [CI] Add smoke test for non-uniform AutoFP8 quantization (#6702) --- tests/quantization/test_fp8.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/quantization/test_fp8.py b/tests/quantization/test_fp8.py index 82dc775f8d812..0602fedf0b8e3 100644 --- a/tests/quantization/test_fp8.py +++ b/tests/quantization/test_fp8.py @@ -13,6 +13,7 @@ from vllm.model_executor.layers.quantization.fp8 import (Fp8KVCacheMethod, MODELS = [ "neuralmagic/Meta-Llama-3-8B-Instruct-FP8-KV", "nm-testing/Phi-3-mini-128k-instruct-FP8", + "nm-testing/Qwen2-0.5B-Instruct-FP8-SkipQKV", ]