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 14, 05:30 UTC
[Misc] Add removed encoder-decoder models to previously supported models list (#24961)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Isotr0py committed
Sep 17, 2025 at 01:46 UTC
d593cf28fa020dbae53dae19122aac8aeeeae0bc
1 file changed
+11
-1
vllm/model_executor/models/registry.py
+11
-1
@@ -319,7 +319,17 @@ _SUBPROCESS_COMMAND = [
319
sys.executable, "-m", "vllm.model_executor.models.registry"
320
]
321
322
-_PREVIOUSLY_SUPPORTED_MODELS = {"Phi3SmallForCausalLM": "0.9.2"}
322
+_PREVIOUSLY_SUPPORTED_MODELS = {
323
+ "Phi3SmallForCausalLM": "0.9.2",
324
+ # encoder-decoder models except whisper
325
+ # have been removed for V0 deprecation.
326
+ "BartModel": "0.10.2",
327
+ "BartForConditionalGeneration": "0.10.2",
328
+ "DonutForConditionalGeneration": "0.10.2",
329
+ "Florence2ForConditionalGeneration": "0.10.2",
330
+ "MBartForConditionalGeneration": "0.10.2",
331
+ "MllamaForConditionalGeneration": "0.10.2",
332
+}
333
334
335
@dataclass(frozen=True)