This commit is contained in:
2026-03-29 17:41:28 +08:00
parent 7ff076530a
commit d84cdc00e5
+7 -7
View File
@@ -24,7 +24,7 @@
"ctx": "24576",
"trust_remote": true,
"valid_tp": [2],
"max_num_seqs": "32",
"max_num_seqs": "8",
"max_tokens": "16384",
"gpu_util": "0.98",
"enforce_eager": false,
@@ -41,7 +41,7 @@
"ctx": "32768",
"trust_remote": true,
"valid_tp": [1, 2],
"max_num_seqs": "64",
"max_num_seqs": "8",
"max_tokens": "32768",
"gpu_util": "0.98",
"tool_call_parser": "qwen3_xml",
@@ -54,7 +54,7 @@
"ctx": "65536",
"trust_remote": true,
"valid_tp": [1, 2],
"max_num_seqs": "64",
"max_num_seqs": "8",
"max_tokens": "32768",
"gpu_util": "0.98",
"tool_call_parser": "qwen3_xml",
@@ -69,9 +69,9 @@
"ctx": "65536",
"trust_remote": true,
"valid_tp": [1, 2],
"max_num_seqs": "64",
"max_tokens": "32768",
"gpu_util": "0.98",
"max_num_seqs": "8",
"max_tokens": "4096",
"gpu_util": "0.92",
"tool_call_parser": "qwen3_xml",
"enable_auto_tool_choice": true,
"served_model_name": "Qwen3.5-35B-A3B-GPTQ-Int4",
@@ -83,7 +83,7 @@
"ctx": "65536",
"trust_remote": true,
"valid_tp": [1, 2],
"max_num_seqs": "64",
"max_num_seqs": "8",
"max_tokens": "32768",
"gpu_util": "0.98",
"tool_call_parser": "qwen3_xml",