benchgap
Other · model

Inkling 2 Mini benchmark scores

As of 2026-11-21, Inkling 2 Mini (Other) has measured scores on 30 benchmarks and estimated scores on 121 more.

BenchmarkScoreSource
AA-GPQA Diamond87.8%estimated ± 2.7 pp, high confidence
AA-HLE27.7%estimated ± 3.9 pp, high confidence
AA-SciCode56.6%estimated ± 3.5 pp, medium confidence
Artificial Analysis Intelligence Index20.1%measured
CritPt2.8%estimated ± 4.0 pp, high confidence
GDPval-AA47.0%estimated ± 6.2 pp, medium confidence
GPQA Diamond86.6%measured
HLE39.5%estimated ± 6.5 pp, medium confidence
AA-LCR79.5%estimated ± 12.5 pp, low confidence
AA-Omniscience Accuracy20.5%measured
SciCode47.2%estimated ± 3.6 pp, high confidence
Terminal-Bench 2.171.6%measured
BrowseComp79.0%measured
HLE w/ tools52.3%estimated ± 5.6 pp, medium confidence
LongBench v266.9%measured
VITA-Bench44.5%estimated ± 6.7 pp, low confidence
τ²-bench results93.0%estimated ± 11.8 pp, low confidence
AA Coding Index74.3%estimated ± 2.3 pp, low confidence
AA-MMMU-Pro74.1%estimated ± 1.5 pp, high confidence
APEX-Agents42.2%estimated ± 12.6 pp, low confidence
APEX-Agents-AA38.1%estimated ± 3.6 pp, medium confidence
BioMysteryBench (human-difficult)31.2%estimated ± 7.8 pp, low confidence
DeepSearchQA85.3%measured
SWE-bench Verified79.2%measured
BFCL v476.7%estimated ± 4.7 pp, low confidence
CyberGym70.2%estimated ± 7.3 pp, medium confidence
JobBench49.6%measured
SkillsBench50.3%estimated ± 4.1 pp, medium confidence
SWE-bench Pro57.6%estimated ± 4.5 pp, high confidence
WideResearch81.2%estimated ± 1.2 pp, medium confidence
AA Agentic Index32.8%estimated ± 8.2 pp, medium confidence
MMLU-Pro85.2%estimated ± 2.5 pp, medium confidence
GPQA86.6%measured
MMLU88.1%estimated ± 0.9 pp, high confidence
Vals GPQA Diamond85.9%estimated ± 2.1 pp, high confidence
Vals LiveCodeBench80.7%estimated ± 10.0 pp, low confidence
Vals MMLU-Pro85.7%estimated ± 0.9 pp, medium confidence
MMMLU88.5%estimated ± 1.1 pp, medium confidence
MMMU82.6%estimated ± 1.7 pp, high confidence
AA MMLU-Pro84.2%estimated ± 0.8 pp, medium confidence
Gert Labs61.4%estimated ± 7.1 pp, medium confidence
AA-AnalystAgent46.3%measured
AA EnterpriseOps-Gym46.1%estimated ± 1.2 pp, medium confidence
AA Harvey LAB v1.090.3%estimated ± 9.8 pp, low confidence
AA Tau3 Banking34.9%estimated ± 3.6 pp, medium confidence
AA Terminal-Bench 2.181.7%estimated ± 2.6 pp, medium confidence
ApprenticeBench15.5%estimated ± 6.0 pp, low confidence
ARC-AGI-183.7%estimated ± 2.5 pp, medium confidence
ARC-AGI-238.4%estimated ± 9.7 pp, low confidence
CursorBench 3.260.2%estimated ± 4.1 pp, low confidence
Terminal-Bench 3.08.7%estimated ± 9.1 pp, low confidence
FrontierCode 1.1 Main37.2%estimated ± 3.5 pp, low confidence
FrontierSWE v225.8%estimated ± 8.0 pp, low confidence
MazeBench0.0%estimated ± 5.1 pp, medium confidence
OfficeQA Pro60.5%estimated ± 2.9 pp, low confidence
OSWorld-Verified71.5%estimated ± 5.7 pp, medium confidence
ParseBench-161.8%estimated ± 1.0 pp, medium confidence
Terminal-Bench 4.113.2%estimated ± 4.6 pp, medium confidence
Vals SWE-bench75.1%estimated ± 6.6 pp, medium confidence
Terminal-Bench 2.1 (Vals AI)61.2%estimated ± 7.3 pp, medium confidence
CWE-Bench-AA50.2%estimated ± 6.4 pp, medium confidence
AA Cyber Index38.1%estimated ± 13.5 pp, low confidence
DeepsecBench-AA12.6%estimated ± 9.0 pp, low confidence
Harvey LAB-AA v1.17.4%estimated ± 4.0 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)51.9%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)55.5%estimated ± 1.1 pp, low confidence
AA ITBench45.3%estimated ± 4.1 pp, high confidence
Bug Hunt Bench20.7%estimated ± 8.9 pp, low confidence
Chartography (no tools)22.1%measured
Chartography (tools)72.5%measured
CursorBench 4.046.5%estimated ± 3.3 pp, medium confidence
DeepSWE48.2%measured
DRACO76.4%estimated ± 2.9 pp, medium confidence
FrontierCode 2.015.1%measured
HLE w/o tools30.7%measured
OSWorld 2.1 Offline59.6%estimated ± 7.9 pp, low confidence
PostTrainBench v1.120.1%estimated ± 4.5 pp, medium confidence
SWE Multilingual73.7%measured
SWE-bench Multimodal56.1%estimated ± 8.3 pp, low confidence
Terminal-Bench 4.05.6%measured
Terminal-Bench-Science 0.11.4%measured
Toolathlon-Verified76.6%estimated ± 2.8 pp, low confidence
VulcanBench Frontier v469.8%estimated ± 5.0 pp, high confidence
GDP.pdf16.5%estimated ± 4.1 pp, medium confidence
AA Terminal-Bench 4.04.3%estimated ± 6.3 pp, low confidence
ARC-AGI-30.7%estimated ± 7.2 pp, low confidence
cursorBench4134.0%estimated ± 6.0 pp, low confidence
CWE-bench57.5%estimated ± 4.6 pp, high confidence
MKA43.1%estimated ± 5.6 pp, low confidence
Vibe Code Bench36.9%estimated ± 8.4 pp, medium confidence
ParseBench-264.7%estimated ± 5.0 pp, medium confidence
CharXiv88.3%measured
CharXiv w/o tools80.6%measured
ExploitGym0.0%estimated ± 9.2 pp, low confidence
AI2D_TEST93.5%estimated ± 0.9 pp, low confidence
C-Eval91.6%estimated ± 1.0 pp, medium confidence
Claw-Eval70.9%estimated ± 4.5 pp, medium confidence
CountBench97.3%estimated ± 1.3 pp, low confidence
DeepPlanning51.3%estimated ± 13.1 pp, low confidence
DynaMath87.7%estimated ± 0.8 pp, low confidence
ERQA63.8%estimated ± 2.9 pp, medium confidence
INCLUDE84.7%estimated ± 2.0 pp, medium confidence
MathVision89.0%measured
MCP Atlas82.7%measured
MLVU (M-Avg)87.0%estimated ± 1.5 pp, high confidence
MMLU-ProX84.4%estimated ± 1.3 pp, high confidence
MMLU-Redux93.6%estimated ± 0.8 pp, medium confidence
MMMU-Pro74.9%measured
NL2Repo51.5%estimated ± 6.0 pp, medium confidence
OmniDocBench 1.590.2%estimated ± 2.2 pp, high confidence
PostTrainBench v118.0%estimated ± 5.5 pp, low confidence
QwenClawBench68.1%estimated ± 2.6 pp, medium confidence
RealWorldQA85.4%estimated ± 1.4 pp, medium confidence
ScreenSpot Pro76.5%measured
SimpleVQA72.6%estimated ± 7.3 pp, medium confidence
SuperGPQA70.5%estimated ± 8.5 pp, low confidence
Toolathlon48.5%estimated ± 4.2 pp, high confidence
V*95.3%estimated ± 6.5 pp, low confidence
Video-MME (w/o subtitle)82.8%estimated ± 1.3 pp, high confidence
Video-MME (with subtitle)88.1%estimated ± 1.9 pp, high confidence
VideoMMMU86.8%estimated ± 1.5 pp, medium confidence
We-Math87.5%estimated ± 2.1 pp, low confidence
MedXpertQA (MM)63.0%estimated ± 2.4 pp, high confidence
MedXpertQA (Text)45.0%estimated ± 4.5 pp, low confidence
MMLU-Pro (Arcee)85.7%estimated ± 3.4 pp, medium confidence
React Native Evals79.3%estimated ± 2.7 pp, medium confidence
ResearchClawBench19.6%estimated ± 1.8 pp, high confidence
SWE-Rebench61.7%estimated ± 3.6 pp, medium confidence
PostTrainBench30.7%estimated ± 4.5 pp, medium confidence
HealthBench (length-adjusted)54.1%estimated ± 3.7 pp, medium confidence
HealthBench Professional40.8%estimated ± 4.2 pp, low confidence
HLE-Verified15.9%estimated ± 7.1 pp, low confidence
ScarfBench61.2%estimated ± 3.8 pp, low confidence
STEMBench2.3%estimated ± 0.2 pp, low confidence
SimpleQA18.8%estimated ± 4.3 pp, medium confidence
LiveCodeBench78.8%estimated ± 14.6 pp, low confidence
Agents' Last Exam26.6%estimated ± 11.6 pp, low confidence
BabyVision w/ Python88.8%measured
ZeroBench w/ Python34.0%measured
BabyVision53.4%measured
MMVU75.3%estimated ± 2.7 pp, medium confidence
PerceptionBench57.5%measured
ZeroBench17.0%measured
LVBench70.9%estimated ± 8.2 pp, low confidence
OCRBench V270.3%estimated ± 3.3 pp, high confidence
BioBasicsBench34.1%estimated ± 3.8 pp, low confidence
LifeVal70.4%estimated ± 1.6 pp, medium confidence
MMMU-Pro w/ Python78.6%estimated ± 2.6 pp, high confidence
MathVision w/ Python94.6%estimated ± 0.5 pp, low confidence
PinchBench90.2%estimated ± 3.6 pp, medium confidence
SWE-Atlas Codebase QnA42.7%estimated ± 4.8 pp, high confidence