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 7, 11:30 UTC
[CI/Build] Enable Qwen3.5 tests on CI (#35763)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Isotr0py committed
Mar 3, 2026 at 01:43 UTC
cc0d565f40814b7406ac7a420725c54ce6ebd116
1 file changed
+6
-10
tests/models/registry.py
+6
-10
@@ -1005,24 +1005,20 @@ _MULTIMODAL_EXAMPLE_MODELS = {
1005
min_transformers_version="4.57",
1006
),
1007
"Qwen3_5ForConditionalGeneration": _HfExamplesInfo(
1008
- "Qwen/Qwen3.5-9B-Instruct",
1008
+ "Qwen/Qwen3.5-0.8B",
1009
max_model_len=4096,
1010
- min_transformers_version="5.1.0",
1010
),
1011
"Qwen3_5MoeForConditionalGeneration": _HfExamplesInfo(
1013
- "Qwen/Qwen3.5-35B-A3B-Instruct",
1012
+ "Qwen/Qwen3.5-35B-A3B",
1013
max_model_len=4096,
1015
- min_transformers_version="5.1.0",
1014
),
1015
"Qwen3_5MTP": _HfExamplesInfo(
1018
- "Qwen/Qwen3.5-9B-Instruct",
1019
- speculative_model="Qwen/Qwen3.5-9B-Instruct",
1020
- min_transformers_version="5.1.0",
1016
+ "Qwen/Qwen3.5-0.8B",
1017
+ speculative_model="Qwen/Qwen3.5-0.8B",
1018
),
1019
"Qwen3_5MoeMTP": _HfExamplesInfo(
1023
- "Qwen/Qwen3.5-35B-A3B-Instruct",
1024
- speculative_model="Qwen/Qwen3.5-35B-A3B-Instruct",
1025
- min_transformers_version="5.1.0",
1020
+ "Qwen/Qwen3.5-35B-A3B",
1021
+ speculative_model="Qwen/Qwen3.5-35B-A3B",
1022
),
1023
"Qwen3OmniMoeForConditionalGeneration": _HfExamplesInfo(
1024
"Qwen/Qwen3-Omni-30B-A3B-Instruct",