Spaces:

taagarwa
/

coding-agent-leaderboard

Running

taagarwa commited on 29 days ago

Commit

343cf48

1 Parent(s): 835297b

🐛 Specify thinking level for gpt 5.5

Files changed (2) hide show

results/swe-bench-pro--ansible-gpt-5-5-codex.json CHANGED Viewed

@@ -12,8 +12,8 @@
         "url": "https://github.com/openai/codex"
     },
     "model": {
-        "name": "GPT 5.5",
-        "repo": "GPT 5.5",
         "is_oss": true,
         "num_params": 9700,
         "precision": "bf16",

         "url": "https://github.com/openai/codex"
     },
     "model": {
+        "name": "GPT 5.5 - high",
+        "repo": "GPT 5.5 - high",
         "is_oss": true,
         "num_params": 9700,
         "precision": "bf16",

results/swe-bench-verified-gpt-5-5-codex.json CHANGED Viewed

@@ -12,8 +12,8 @@
         "url": "https://github.com/openai/codex"
     },
     "model": {
-        "name": "GPT 5.5",
-        "repo": "GPT 5.5",
         "is_oss": true,
         "num_params": 9700,
         "precision": "bf16",

         "url": "https://github.com/openai/codex"
     },
     "model": {
+        "name": "GPT 5.5 - high",
+        "repo": "GPT 5.5 - high",
         "is_oss": true,
         "num_params": 9700,
         "precision": "bf16",