[Eagle] Refactor eagle speculative decoding (#3986)

Co-authored-by: Ke Bao <ISPObaoke@163.com>
This commit is contained in:
Ying Sheng
2025-03-05 08:06:07 -08:00
committed by GitHub
parent 5be8f1ed98
commit d3d4d76758
22 changed files with 670 additions and 352 deletions

View File

@@ -100,7 +100,7 @@ class TpModelWorkerClient:
def get_memory_pool(self):
return (
self.worker.model_runner.req_to_token_pool,
self.worker.model_runner.token_to_kv_pool,
self.worker.model_runner.token_to_kv_pool_allocator,
)
def forward_thread_func(self):