benchgap
Other · model

BTL-3 benchmark scores

As of 2026-10-07, BTL-3 (Other) has measured scores on 2 benchmarks and estimated scores on 9 more.

BenchmarkScoreSource
AA-SciCode45.9%estimated ± 3.7 pp, high confidence
GDPval-AA100.0%estimated ± 5.8 pp, low confidence
LiveCodeBench v688.1%measured
SciCode49.8%estimated ± 5.4 pp, medium confidence
τ²-bench results100.0%estimated ± 14.1 pp, low confidence
AA Coding Index57.4%estimated ± 6.4 pp, medium confidence
SWE-bench Verified72.2%estimated ± 10.5 pp, low confidence
BFCL v488.5%measured
SWE-bench Pro58.3%estimated ± 4.7 pp, high confidence
Claw-Eval64.9%estimated ± 2.3 pp, low confidence
NL2Repo45.6%estimated ± 3.1 pp, low confidence