benchgap
Other · model

Agents-A1 benchmark scores

As of 2026-11-21, Agents-A1 (Other) has measured scores on 6 benchmarks and estimated scores on 51 more.

BenchmarkScoreSource
AA-GPQA Diamond89.1%estimated ± 7.1 pp, medium confidence
AA-HLE40.2%estimated ± 6.5 pp, medium confidence
Artificial Analysis Intelligence Index33.0%estimated ± 10.7 pp, low confidence
CritPt14.8%estimated ± 6.1 pp, medium confidence
GDPval-AA34.4%estimated ± 7.7 pp, medium confidence
GPQA Diamond91.8%estimated ± 2.3 pp, high confidence
HLE47.6%measured
IFBench74.3%estimated ± 9.2 pp, medium confidence
AA-LCR68.1%estimated ± 12.5 pp, low confidence
AA-Omniscience Accuracy39.5%estimated ± 9.1 pp, medium confidence
BrowseComp75.5%measured
HLE w/ tools47.6%measured
IFEval94.8%measured
LongBench v260.2%measured
VITA-Bench38.8%measured
τ²-bench results95.8%estimated ± 3.1 pp, high confidence
APEX-Agents-AA21.5%estimated ± 4.4 pp, high confidence
DeepSearchQA82.7%estimated ± 6.8 pp, low confidence
CyberGym65.2%estimated ± 7.3 pp, medium confidence
JobBench36.9%estimated ± 10.7 pp, low confidence
SkillsBench46.7%estimated ± 4.1 pp, medium confidence
WideResearch75.6%estimated ± 3.5 pp, high confidence
AA Agentic Index29.2%estimated ± 8.2 pp, medium confidence
AA-IFBench79.6%estimated ± 8.8 pp, medium confidence
GPQA91.7%estimated ± 1.9 pp, high confidence
Vals MMLU-Pro87.4%estimated ± 2.4 pp, medium confidence
Gert Labs57.0%estimated ± 7.1 pp, medium confidence
AA Harvey LAB v1.086.8%estimated ± 1.7 pp, medium confidence
AA Tau3 Banking32.5%estimated ± 3.6 pp, medium confidence
ARC-AGI-186.4%estimated ± 11.6 pp, low confidence
MazeBench0.0%estimated ± 6.4 pp, low confidence
OSWorld-Verified67.6%estimated ± 5.7 pp, medium confidence
CWE-Bench-AA47.5%estimated ± 5.2 pp, medium confidence
AA Cyber Index32.4%estimated ± 12.8 pp, low confidence
DeepsecBench-AA9.8%estimated ± 9.0 pp, low confidence
AA ITBench43.8%estimated ± 4.1 pp, high confidence
DRACO73.4%estimated ± 2.9 pp, medium confidence
HLE w/o tools40.1%estimated ± 6.0 pp, medium confidence
OSWorld 2.162.7%estimated ± 7.2 pp, low confidence
Toolathlon-Verified63.7%estimated ± 5.7 pp, medium confidence
AA AutomationBench41.9%estimated ± 7.9 pp, low confidence
GDP.pdf7.7%estimated ± 6.1 pp, low confidence
ARC-AGI-30.0%estimated ± 8.3 pp, low confidence
Hermes Bench67.0%estimated ± 1.8 pp, medium confidence
Hermes Index25.9%estimated ± 9.3 pp, low confidence
ExploitGym0.0%estimated ± 9.2 pp, low confidence
Claw-Eval62.7%estimated ± 6.4 pp, medium confidence
DeepPlanning26.4%estimated ± 9.1 pp, low confidence
MCP Atlas75.3%estimated ± 10.3 pp, low confidence
MMLU-Redux96.5%estimated ± 2.1 pp, medium confidence
QwenClawBench52.0%estimated ± 2.0 pp, medium confidence
SuperGPQA87.0%estimated ± 7.2 pp, medium confidence
Toolathlon45.0%estimated ± 4.2 pp, high confidence
ResearchClawBench17.4%estimated ± 2.3 pp, medium confidence
Agents' Last Exam23.1%estimated ± 11.6 pp, low confidence
LifeVal66.0%estimated ± 1.6 pp, medium confidence
PinchBench90.0%estimated ± 3.6 pp, medium confidence