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 8, 01:15 UTC
[Bugfix] Add Quant Config to Llava Next Projector (#34847)
Signed-off-by: Alex Brooks <albrooks@redhat.com>
Alex Brooks committed
Feb 19, 2026 at 00:18 UTC
4611af1663e268b5a64221c999868779632296a7
1 file changed
+2
vllm/model_executor/models/llava_next.py
+2
@@ -285,6 +285,8 @@ class LlavaNextForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsP
285
text_hidden_size=config.text_config.hidden_size,
286
projector_hidden_act=config.projector_hidden_act,
287
multimodal_projector_bias=config.multimodal_projector_bias,
288
+ quant_config=quant_config,
289
+ prefix=maybe_prefix(prefix, "multi_modal_projector"),
290
)
291
292
with self._mark_language_model(vllm_config):