Skip to content

update max_model_len to plus 1 to adjust vllm change #793

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

update max_model_len to plus 1 to adjust vllm change #793

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion tpu_commons/runner/tpu_jax_runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -206,7 +206,10 @@ def _init_inputs(self) -> None:

self.sliding_window = model_config.get_sliding_window()
self.block_size = cache_config.block_size
self.max_model_len = model_config.max_model_len
# Note: https://github.com/vllm-project/vllm/pull/26168 allows
# prefill_length to be max-model-len, and decode for 1 token.
# so the total sequence length becomes max_model_len + 1.
self.max_model_len = model_config.max_model_len + 1
self.max_num_blocks_per_req = cdiv(self.max_model_len, self.block_size)
# InputBatch needs to work with sampling tensors greater than padding
# to avoid dynamic shapes. Also, avoid suboptimal alignment.
Expand Down
Loading