Fix Weight loading for Qwen3.5-MTP and Qwen3-VL using runai_streamer (#42716)

Signed-off-by: weizhoublue <weizhou.lan@daocloud.io>
This commit is contained in:
weizhoublue
2026-05-16 17:54:27 -07:00
committed by GitHub
parent 0867497368
commit a94189295b
2 changed files with 4 additions and 4 deletions
+2 -2
View File
@@ -175,8 +175,8 @@ class Qwen3_5MultiTokenPredictor(nn.Module):
param,
curr_expert_weight,
name,
shard_id,
expert_id,
shard_id=shard_id,
expert_id=expert_id,
return_success=True,
)
if success:
+2 -2
View File
@@ -152,8 +152,8 @@ class Qwen3MoeLLMModel(Qwen3MoeModel):
param,
curr_expert_weight,
name,
shard_id,
expert_id,
shard_id=shard_id,
expert_id=expert_id,
return_success=True,
)
if success: