Complete config: all fields included (env, limit, limit_per_task, judge, judge_url, repeats, max_input_tokens, max_turns, concurrency)

Co-Authored-By: Claude <noreply@anthropic.com>

evalharness/config/dp4-nothink.yaml

@@ -3,6 +3,7 @@ default:
   top_p: 1.0
   stream: true
   max_tokens: 32768
+  concurrency: 8
 
 aime24:
   temperature: 1.0
@@ -22,54 +23,98 @@ hmmt26:
 
 imo_answerbench:
   temperature: 1.0
+  limit_per_task: 25
+  judge: dp4-flash
+  judge_url: http://174.1.51.4:30000/v1
+
+gsm8k:
+  limit: 200
+
+competition_math:
+  limit_per_task: 40
+
+mmlu:
+  max_tokens: 8192
+  limit_per_task: 4
+
+cmmlu:
+  max_tokens: 8192
+  limit_per_task: 3
+
+mmlu_pro:
+  max_tokens: 8192
+  limit_per_task: 20
+
+arc:
+  max_tokens: 8192
+  limit: 200
+
+hellaswag:
+  max_tokens: 8192
+  limit: 400
+
+winogrande:
+  max_tokens: 8192
+  limit: 200
+
+bbh:
+  limit_per_task: 10
 
 gpqa_diamond:
   temperature: 1.0
   max_tokens: 8192
-
+  limit: 200
-mmlu:
-  max_tokens: 8192
-
-mmlu_pro:
-  max_tokens: 8192
-
-cmmlu:
-  max_tokens: 8192
-
-arc:
-  max_tokens: 8192
-
-hellaswag:
-  max_tokens: 8192
-
-winogrande:
-  max_tokens: 8192
 
 simple_qa:
   max_tokens: 8192
+  limit: 200
 
 trivia_qa:
   max_tokens: 8192
+  limit: 200
+
+drop:
+  limit: 200
+
+hle:
+  limit_per_task: 25
+  judge: dp4-flash
+  judge_url: http://174.1.51.4:30000/v1
 
 humaneval:
   temperature: 1.0
 
+bigcodebench:
+  limit: 200
+
 live_code_bench:
   temperature: 1.0
+  limit: 200
 
 longbench_v2:
   max_tokens: 8192
+  limit_per_task: 66
   max_input_tokens: 128000
 
 openai_mrcr:
   max_tokens: 8192
+  limit: 200
   max_input_tokens: 128000
 
 bfcl_v3:
   max_tokens: 4096
+  limit_per_task: 20
+  env: bfcl_mock
 
 general_fc:
   max_tokens: 4096
+  limit: 400
 
 tau2_bench:
   max_tokens: 16384
+  env: tau2_official
+  max_turns: 50
+
+swe_bench_verified:
+  limit: 70
+  concurrency: 4