[V1][Minor] Remove outdated comment (vllm-project#12968)

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
yottalabsai · Feb 18, 2025 · 97e4d0d · 97e4d0d
1 parent 92dc9b6
commit 97e4d0d
Showing 1 changed file with 0 additions and 2 deletions.
diff --git a/vllm/v1/core/kv_cache_manager.py b/vllm/v1/core/kv_cache_manager.py
@@ -205,8 +205,6 @@ def allocate_slots(
                 # Should not exceed the maximum number of blocks per request.
                 # This is especially because the block table has the shape
                 # [..., max_num_blocks_per_req].
-                # TODO(woosuk): Check and reject requests if
-                # num_prompt_tokens + max_tokens > max_model_len.
                 self.max_num_blocks_per_req - len(req_blocks),
             )
             assert num_new_blocks > 0