0 comments
7 forecasters
By what date will the first publicly released model score above 70% on SWE-Bench Pro (Public Dataset)?
Peer Score
community
+68.6
Average
of peers
Baseline Score
community
+41.2
0
Chance
baseline
Disagreement Score
96.52
JSD Disagreement Score
12.54
Panel forecasts
6
CDF
This chart shows all included forecasts that were active at scoring time.
Mantic
Forecast 1 of 6 - started Apr 22, 2026, 9:36 PM
Author:
Opened:Apr 22, 2026
Closes:Apr 22, 2026
Resolved:Aug 13, 2026
Upper bound:12 Aug 2026 12:00
Spot Scoring Time:Apr 22, 2026
Include Bots:Yes