[Eagle] Refactor eagle speculative decoding (#3986)
Co-authored-by: Ke Bao <ISPObaoke@163.com>
This commit is contained in:
@@ -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):
|
||||
|
||||
Reference in New Issue
Block a user