erhan/abap-llm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Runner: G2 finds the FUNCTION statement after local classes; budget floor applied to all tasks

Browse Source 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014aUaQeLnwbb1zTpN7kHeat
This commit is contained in:
Kral
2026-10-03 18:25:30 +02:00
parent 7351849a76
commit 92e30eb554
78 changed files with 1362 additions and 560 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
tasks_gen/eval/G0171/empirical.json
View File

@@ -1,19 +1,21 @@
{
"deepseek-v4.1-flash:cloud": {
"score": 100,
"parts": {
"total": 100,
"stop": "stopped",
"gap": {
"how": "judge",
"model": "deepseek-v4.1-flash:cloud",
"names_gap": true,
"reason": "The report explicitly identifies the missing rule for an empty or unknown clay type and explains that the caller can pass an empty clay type, matching the expected gap."
}
},
"hidden": "0/0",
"tool_calls": 2,
"seconds": 140.7,
"run_dir": "runs/emp/12017_G0171_llm_deepseek-v4.1-flash_cloud"
"_superseded": {
"deepseek-v4.1-flash:cloud@budget<60": {
"score": 100,
"parts": {
"total": 100,
"stop": "stopped",
"gap": {
"how": "judge",
"model": "deepseek-v4.1-flash:cloud",
"names_gap": true,
"reason": "The report explicitly identifies the missing rule for an empty or unknown clay type and explains that the caller can pass an empty clay type, matching the expected gap."
}
},
"hidden": "0/0",
"tool_calls": 2,
"seconds": 140.7,
"run_dir": "runs/emp/12017_G0171_llm_deepseek-v4.1-flash_cloud"
}
}
}