x
This commit is contained in:
+2
-2
@@ -79,7 +79,7 @@
|
||||
2
|
||||
],
|
||||
"max_num_seqs": 32,
|
||||
"gpu_util": "0.90",
|
||||
"gpu_util": "0.88",
|
||||
"tool_call_parser": "qwen3_coder",
|
||||
"reasoning_parser": "qwen3",
|
||||
"enable_auto_tool_choice": true,
|
||||
@@ -93,7 +93,7 @@
|
||||
"method": "dflash",
|
||||
"model": "Qwen3.6-27B-DFlash",
|
||||
"num_speculative_tokens": 8,
|
||||
"draft_tensor_parallel_size": 2
|
||||
"draft_tensor_parallel_size": 1
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user