[Model Runner V2] Move MM encoder to Model States [3/N] (#35564)

Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
This commit is contained in:
Woosuk Kwon
2026-02-27 18:32:38 -08:00
committed by GitHub
parent 86ac7bcf84
commit 1a014a0a93
7 changed files with 135 additions and 111 deletions
@@ -44,7 +44,6 @@ class EagleSpeculator:
# the draft model's hidden size can be different from the target model's
# hidden size (e.g., Llama 3.3 70B).
self.hidden_size = self.draft_model_config.get_hidden_size()
self.inputs_embeds_size = self.draft_model_config.get_inputs_embeds_size()
self.vocab_size = self.draft_model_config.get_vocab_size()
self.dtype = vllm_config.model_config.dtype