Skip to content

GitLab

Commit c7b0d0d4 authored by zhuwenwen's avatar zhuwenwen
Browse files

Merge remote-tracking branch 'origin/v0.9.2-dev-fix' into v0.9.2-dev

parents 9d16d5aa e9cfa85e
Show whitespace changes
Inline Side-by-side
Showing with 3 additions and 1 deletion
vllm/v1/worker/gpu_model_runner.py
View file @ c7b0d0d4
...@@ -513,7 +513,9 @@ class GPUModelRunnerBase(LoRAModelRunnerMixin): ...@@ -513,7 +513,9 @@ class GPUModelRunnerBase(LoRAModelRunnerMixin):
# Add the sampled token(s) from the previous step (if any). # Add the sampled token(s) from the previous step (if any).
# This doesn't include "unverified" tokens like spec tokens. # This doesn't include "unverified" tokens like spec tokens.
num_new_tokens = len(new_token_ids) num_new_tokens = len(new_token_ids)
if num_new_tokens > 0: if num_new_tokens == 1:
req_state.output_token_ids.append(new_token_ids[-1])
elif num_new_tokens > 0:
req_state.output_token_ids.extend( req_state.output_token_ids.extend(
new_token_ids) new_token_ids)
if len(spec_token_ids) > 0: if len(spec_token_ids) > 0:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment