benchgap
Other · model

Atria Dawn Preview benchmark scores

As of 2026-11-22, Atria Dawn Preview (Other) has measured scores on 11 benchmarks and estimated scores on 58 more.

BenchmarkScoreSource
AA-SciCode50.1%estimated ± 3.6 pp, high confidence
GDPval-AA53.2%estimated ± 5.0 pp, high confidence
LiveCodeBench v689.4%estimated ± 2.7 pp, high confidence
SciCode48.6%estimated ± 5.9 pp, medium confidence
Terminal-Bench 2.178.3%measured
BrowseComp92.5%measured
HLE w/ tools63.3%estimated ± 4.9 pp, medium confidence
VITA-Bench47.5%estimated ± 6.7 pp, low confidence
τ²-bench results93.7%estimated ± 11.8 pp, low confidence
AA Coding Index64.8%estimated ± 7.5 pp, medium confidence
APEX-Agents69.4%estimated ± 7.4 pp, low confidence
APEX-Agents-AA38.3%estimated ± 3.6 pp, medium confidence
DeepSearchQA96.0%measured
SWE-bench Verified80.4%estimated ± 3.6 pp, high confidence
AutomationBench53.8%measured
BFCL v477.0%measured
CyberGym86.5%measured
JobBench50.3%measured
SkillsBench66.4%measured
SWE-bench Pro59.6%measured
τ³-bench results41.2%measured
WideResearch81.9%measured
AA Agentic Index52.5%estimated ± 5.8 pp, low confidence
Vals LiveCodeBench84.5%estimated ± 4.7 pp, medium confidence
Gert Labs53.6%estimated ± 5.3 pp, medium confidence
AA EnterpriseOps-Gym46.1%estimated ± 1.2 pp, medium confidence
AA Harvey LAB v1.093.5%estimated ± 9.8 pp, low confidence
AA Tau3 Banking100.0%estimated ± 3.3 pp, low confidence
AA Terminal-Bench 2.181.7%estimated ± 2.6 pp, high confidence
ApprenticeBench30.5%estimated ± 4.0 pp, medium confidence
CursorBench 3.154.4%estimated ± 5.6 pp, low confidence
CursorBench 3.257.6%estimated ± 3.6 pp, high confidence
Terminal-Bench 3.012.2%estimated ± 9.1 pp, low confidence
FrontierCode 1.1 Main39.5%estimated ± 5.0 pp, high confidence
FrontierSWE v214.9%estimated ± 8.3 pp, medium confidence
MazeBench0.0%estimated ± 5.1 pp, medium confidence
OSWorld-Verified80.4%estimated ± 4.7 pp, high confidence
Terminal-Bench 4.115.5%estimated ± 10.9 pp, low confidence
Vals SWE-bench81.2%estimated ± 7.2 pp, medium confidence
Terminal-Bench 2.1 (Vals AI)66.3%estimated ± 7.3 pp, medium confidence
CWE-Bench-AA61.9%estimated ± 6.0 pp, low confidence
AA Cyber Index46.3%estimated ± 9.4 pp, medium confidence
DeepsecBench-AA27.8%estimated ± 9.0 pp, low confidence
AA ITBench51.1%estimated ± 4.1 pp, medium confidence
Bug Hunt Bench31.1%estimated ± 12.1 pp, low confidence
CursorBench 4.035.3%estimated ± 3.3 pp, medium confidence
DeepSWE52.3%estimated ± 11.2 pp, low confidence
DRACO88.0%estimated ± 2.9 pp, low confidence
FrontierCode 2.021.0%estimated ± 6.7 pp, medium confidence
PostTrainBench v1.125.4%estimated ± 4.4 pp, high confidence
Toolathlon-Verified74.0%estimated ± 2.2 pp, low confidence
VulcanBench Frontier v479.5%estimated ± 8.3 pp, low confidence
GDP.pdf16.6%estimated ± 4.1 pp, medium confidence
AA Terminal-Bench 4.04.3%estimated ± 11.0 pp, low confidence
ExploitGym33.8%estimated ± 9.2 pp, low confidence
Claw-Eval71.3%estimated ± 4.5 pp, medium confidence
MCP Atlas83.8%estimated ± 6.6 pp, low confidence
MCP-Tasks84.2%estimated ± 6.0 pp, low confidence
NL2Repo46.4%estimated ± 6.1 pp, medium confidence
PostTrainBench v124.8%estimated ± 4.6 pp, high confidence
Toolathlon63.9%estimated ± 4.2 pp, medium confidence
LiveCodeBench Pro83.4%estimated ± 6.0 pp, low confidence
ResearchClawBench20.4%estimated ± 2.3 pp, low confidence
PostTrainBench23.5%estimated ± 3.7 pp, high confidence
FrontierCode 1.1 Extended55.9%estimated ± 3.0 pp, low confidence
Agents' Last Exam30.1%estimated ± 3.1 pp, high confidence
OpenHarmony Bench v1.053.4%estimated ± 4.9 pp, low confidence
LifeVal80.9%estimated ± 1.6 pp, low confidence
PinchBench90.8%estimated ± 3.6 pp, low confidence