benchgap
Other · model

LongCat-2.5 benchmark scores

As of 2026-11-19, LongCat-2.5 (Other) has measured scores on 16 benchmarks and estimated scores on 79 more.

BenchmarkScoreSource
AA-SciCode53.3%estimated ± 3.1 pp, high confidence
GDPval-AA57.6%estimated ± 6.2 pp, medium confidence
LiveCodeBench v695.1%estimated ± 1.9 pp, medium confidence
SciCode57.0%estimated ± 7.6 pp, low confidence
Terminal-Bench 2.188.4%measured
BrowseComp91.9%measured
HLE w/ tools63.1%measured
VITA-Bench47.4%estimated ± 6.7 pp, low confidence
τ²-bench results98.6%estimated ± 3.1 pp, medium confidence
AA Coding Index74.5%estimated ± 2.5 pp, high confidence
AA-MMMU-Pro81.0%estimated ± 4.9 pp, medium confidence
APEX-Agents58.0%estimated ± 7.4 pp, low confidence
APEX-Agents-AA42.6%estimated ± 3.6 pp, low confidence
DeepSearchQA93.5%estimated ± 7.0 pp, low confidence
SWE-bench Verified81.9%estimated ± 2.3 pp, high confidence
AutomationBench51.8%measured
BFCL v476.8%estimated ± 4.7 pp, low confidence
CyberGym87.5%estimated ± 4.6 pp, medium confidence
JobBench65.4%measured
SkillsBench63.3%measured
SWE-bench Pro62.4%estimated ± 3.5 pp, high confidence
WideResearch84.6%estimated ± 1.2 pp, low confidence
AA Agentic Index52.7%estimated ± 8.2 pp, medium confidence
Vals LiveCodeBench77.4%estimated ± 6.1 pp, low confidence
MMMU90.7%estimated ± 3.7 pp, low confidence
Gert Labs79.9%estimated ± 7.1 pp, low confidence
AA EnterpriseOps-Gym46.8%estimated ± 1.2 pp, medium confidence
AA Harvey LAB v1.091.8%estimated ± 1.7 pp, medium confidence
AA Tau3 Banking50.1%estimated ± 3.3 pp, low confidence
AA Terminal-Bench 2.186.2%estimated ± 2.6 pp, high confidence
ApprenticeBench19.5%estimated ± 6.0 pp, low confidence
CursorBench 3.226.6%estimated ± 3.6 pp, low confidence
Terminal-Bench 3.023.7%estimated ± 9.1 pp, low confidence
FrontierCode 1.1 Main46.9%estimated ± 3.5 pp, low confidence
MazeBench0.1%estimated ± 5.1 pp, medium confidence
OSWorld-Verified73.9%estimated ± 3.8 pp, low confidence
ParseBench-169.0%estimated ± 8.6 pp, low confidence
Terminal-Bench 4.132.9%estimated ± 5.0 pp, medium confidence
Vals SWE-bench94.1%estimated ± 5.2 pp, medium confidence
Terminal-Bench 2.1 (Vals AI)74.0%estimated ± 7.3 pp, medium confidence
CWE-Bench-AA60.6%estimated ± 5.2 pp, medium confidence
AA Cyber Index46.3%estimated ± 9.4 pp, medium confidence
DeepsecBench-AA27.0%estimated ± 9.0 pp, low confidence
AA ITBench50.8%estimated ± 4.1 pp, high confidence
Bug Hunt Bench25.9%estimated ± 7.2 pp, low confidence
CursorBench 4.046.0%estimated ± 4.7 pp, high confidence
DeepSWE64.6%measured
DRACO87.5%estimated ± 2.9 pp, low confidence
OSWorld 2.175.4%estimated ± 7.2 pp, low confidence
PostTrainBench v1.132.7%estimated ± 4.5 pp, high confidence
ProgramBench25.5%measured
Terminal-Bench 4.038.5%measured
Terminal-Bench-Science 0.115.2%estimated ± 14.9 pp, low confidence
Toolathlon-Verified76.6%estimated ± 5.7 pp, medium confidence
VulcanBench Frontier v488.5%estimated ± 1.2 pp, low confidence
AA AutomationBench68.0%estimated ± 7.9 pp, low confidence
GDP.pdf21.9%estimated ± 4.1 pp, medium confidence
AA Terminal-Bench 4.036.0%estimated ± 6.3 pp, medium confidence
CWE-bench62.3%estimated ± 11.0 pp, low confidence
MKA40.8%estimated ± 6.6 pp, medium confidence
Hermes Bench70.3%estimated ± 1.5 pp, medium confidence
Hermes Index37.2%estimated ± 7.0 pp, low confidence
CharXiv90.2%measured
CharXiv w/o tools85.1%estimated ± 1.9 pp, low confidence
ExploitGym32.3%estimated ± 9.2 pp, low confidence
AI2D_TEST93.5%estimated ± 0.9 pp, low confidence
Claw-Eval78.3%measured
CountBench98.5%estimated ± 3.7 pp, low confidence
DeepPlanning51.5%estimated ± 13.1 pp, low confidence
DynaMath88.0%estimated ± 0.8 pp, low confidence
ERQA71.2%estimated ± 5.6 pp, medium confidence
MathVision92.5%estimated ± 0.7 pp, medium confidence
MCP Atlas83.1%measured
MLVU (M-Avg)89.0%estimated ± 1.5 pp, high confidence
MMMU-Pro81.7%estimated ± 2.9 pp, high confidence
NL2Repo60.5%measured
QwenClawBench68.5%estimated ± 2.6 pp, medium confidence
RealWorldQA87.4%estimated ± 1.5 pp, high confidence
ScreenSpot Pro86.5%estimated ± 6.8 pp, medium confidence
Toolathlon63.1%estimated ± 4.2 pp, high confidence
Video-MME (with subtitle)90.8%estimated ± 2.9 pp, high confidence
VideoMMMU87.3%estimated ± 1.5 pp, medium confidence
We-Math84.6%estimated ± 7.9 pp, low confidence
MedXpertQA (MM)79.1%estimated ± 6.4 pp, low confidence
ResearchClawBench22.8%estimated ± 1.3 pp, medium confidence
ScarfBench64.4%estimated ± 3.8 pp, medium confidence
Agents' Last Exam38.5%measured
OpenHarmony Bench v1.059.3%estimated ± 2.7 pp, high confidence
BabyVision70.1%estimated ± 10.4 pp, low confidence
MMVU80.7%estimated ± 3.6 pp, medium confidence
SWE-Marathon42.6%estimated ± 12.9 pp, low confidence
LifeVal80.6%estimated ± 1.6 pp, medium confidence
MathVision w/ Python96.5%measured
PinchBench90.8%estimated ± 3.6 pp, low confidence
SWE-Atlas Codebase QnA58.3%measured