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 15, 09:45 UTC
[Misc] Add LoRA code owner (#18387)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Jee Jee Li committed
May 20, 2025 at 18:27 UTC
6b35cb10a0127b7fd9cd23346ac0a4ced72fed3c
2 files changed
+2
-6
.github/CODEOWNERS
+2
@@ -13,6 +13,7 @@
13
/vllm/model_executor/guided_decoding @mgoin @russellb
14
/vllm/multimodal @DarkLight1337 @ywang96
15
/vllm/vllm_flash_attn @LucasWilkinson
16
+/vllm/lora @jeejeelee
17
CMakeLists.txt @tlrmchlsmth
18
19
# vLLM V1
@@ -40,3 +41,4 @@ CMakeLists.txt @tlrmchlsmth
41
/tests/v1/entrypoints/llm/test_struct_output_generate.py @mgoin @russellb
42
/tests/v1/structured_output @mgoin @russellb
43
/tests/weight_loading @mgoin @youkaichao
44
+/tests/lora @jeejeelee
tests/quantization/test_bitsandbytes.py
-6
@@ -37,12 +37,6 @@ models_pre_quant_8bit_to_test = [
37
("yec019/fbopt-350m-8bit", "read pre-quantized 8-bit opt model"),
38
]
39
40
-models_pre_quant_8bit_to_test = [
41
- ('meta-llama/Llama-Guard-3-8B-INT8',
42
- 'read pre-quantized llama 8-bit model'),
43
- ("yec019/fbopt-350m-8bit", "read pre-quantized 8-bit opt model"),
44
-]
45
-
40
41
@pytest.mark.skipif(not is_quant_method_supported("bitsandbytes"),
42
reason='bitsandbytes is not supported on this GPU type.')