benchgap
OpenAI · model

GPT-6.1 Sol benchmark scores

As of 2026-11-21, GPT-6.1 Sol (OpenAI) has measured scores on 55 benchmarks and estimated scores on 113 more.

BenchmarkScoreSource
AA-GPQA Diamond96.0%estimated ± 1.1 pp, medium confidence
AA-HLE52.9%measured
AA-SciCode54.2%measured
AIME2698.2%estimated ± 1.4 pp, medium confidence
Artificial Analysis Intelligence Index54.6%measured
CritPt31.7%measured
GDPval-AA54.8%measured
GPQA Diamond96.2%estimated ± 1.2 pp, medium confidence
HLE69.4%estimated ± 2.2 pp, high confidence
AA-LCR83.0%measured
LiveCodeBench v694.7%estimated ± 4.9 pp, medium confidence
AA-Omniscience Accuracy61.8%measured
SciCode53.9%estimated ± 3.1 pp, high confidence
Terminal-Bench 2.188.9%estimated ± 5.6 pp, low confidence
BrowseComp91.4%estimated ± 3.0 pp, medium confidence
HLE w/ tools62.4%estimated ± 3.1 pp, high confidence
LongBench v268.4%estimated ± 5.5 pp, medium confidence
AA Coding Index78.7%estimated ± 2.3 pp, low confidence
AA-MMMU-Pro86.0%measured
APEX-Agents37.5%estimated ± 7.4 pp, low confidence
APEX-Agents-AA40.0%estimated ± 9.5 pp, medium confidence
BioMysteryBench (human-difficult)49.9%estimated ± 6.8 pp, low confidence
DeepSearchQA89.6%estimated ± 9.6 pp, medium confidence
SWE-bench Verified89.0%estimated ± 5.1 pp, medium confidence
AutomationBench36.1%measured
BFCL v477.0%estimated ± 2.7 pp, medium confidence
CyberGym83.0%estimated ± 4.6 pp, medium confidence
JobBench59.3%estimated ± 8.7 pp, medium confidence
SWE-bench Pro81.4%estimated ± 4.3 pp, high confidence
WideResearch84.3%estimated ± 7.8 pp, low confidence
AA Agentic Index50.6%estimated ± 3.1 pp, high confidence
MMLU-Pro85.5%estimated ± 5.0 pp, low confidence
FailureSensorIQ68.6%estimated ± 1.7 pp, high confidence
GPQA95.8%estimated ± 0.3 pp, medium confidence
Vals GPQA Diamond98.8%estimated ± 1.2 pp, low confidence
Vals LiveCodeBench85.8%estimated ± 5.7 pp, low confidence
Vals MMLU-Pro89.8%estimated ± 1.6 pp, low confidence
MMMLU92.0%estimated ± 2.6 pp, low confidence
MMMU89.9%estimated ± 2.7 pp, medium confidence
Chess Puzzles68.0%measured
AA MMLU-Pro90.5%estimated ± 0.8 pp, medium confidence
Gert Labs73.2%estimated ± 8.5 pp, low confidence
AA-AnalystAgent50.0%measured
AA Tau3 Banking45.7%estimated ± 3.3 pp, medium confidence
ApprenticeBench51.7%estimated ± 12.5 pp, low confidence
ARC-AGI-196.5%measured
ARC-AGI-294.2%measured
BabaIsBench100.0%estimated ± 1.2 pp, high confidence
CursorBench 3.154.4%estimated ± 3.4 pp, medium confidence
CursorBench 3.268.2%estimated ± 3.1 pp, medium confidence
EBR-bench54.3%measured
Terminal-Bench 3.047.1%estimated ± 3.2 pp, medium confidence
FrontierCode 1.1 Main52.4%estimated ± 3.0 pp, medium confidence
FrontierMath v2 (Tier 4)100.0%measured
FrontierMath v2 (Tiers 1-3)93.7%measured
FrontierSWE v256.0%estimated ± 8.0 pp, medium confidence
Furniture Assembly80.0%measured
MaxxBench69.4%estimated ± 4.3 pp, medium confidence
MazeBench17.0%measured
MirrorCode61.1%measured
Mystery Game Puzzles80.0%measured
NYT Connections97.8%measured
OfficeQA Pro67.6%estimated ± 6.8 pp, medium confidence
OSWorld-Verified84.4%estimated ± 8.4 pp, low confidence
ParseBench-169.6%measured
Terminal-Bench 4.160.9%measured
Terminal-Bench Hard69.4%estimated ± 2.8 pp, low confidence
Vals SWE-bench97.4%estimated ± 3.9 pp, medium confidence
Terminal-Bench 2.1 (Vals AI)79.9%estimated ± 8.3 pp, medium confidence
VulcanBench v387.6%estimated ± 3.9 pp, low confidence
CWE-Bench-AA63.3%estimated ± 3.8 pp, high confidence
AA Cyber Index30.4%estimated ± 5.3 pp, low confidence
DeepsecBench-AA30.8%estimated ± 7.5 pp, medium confidence
Harvey LAB-AA v1.16.9%measured
HealthBench Professional (length-adjusted, Artificial Analysis)62.5%estimated ± 4.3 pp, medium confidence
HealthBench Professional (raw, Artificial Analysis)70.4%estimated ± 0.8 pp, medium confidence
AA ITBench47.6%estimated ± 5.1 pp, low confidence
Bug Hunt Bench44.3%measured
Chartography (tools)86.9%estimated ± 4.1 pp, low confidence
CursorBench 4.054.7%estimated ± 3.3 pp, high confidence
CWE-bench v163.9%estimated ± 4.5 pp, medium confidence
DeepSWE71.9%measured
DRACO87.9%estimated ± 0.7 pp, low confidence
FrontierCode 2.039.6%measured
FrontierPhysics19.6%measured
HLE w/o tools59.6%estimated ± 4.9 pp, high confidence
OSWorld 2.178.8%estimated ± 5.3 pp, medium confidence
OSWorld 2.1 Offline70.6%estimated ± 1.9 pp, medium confidence
PostTrainBench v1.141.9%estimated ± 2.2 pp, high confidence
SpireClimb23.8%measured
StealthMark67.8%measured
SWE-bench Multimodal56.7%estimated ± 5.6 pp, low confidence
Terminal-Bench 4.062.0%estimated ± 4.5 pp, high confidence
Terminal-Bench-Science 0.157.0%measured
Toolathlon-Verified77.9%estimated ± 1.8 pp, medium confidence
VulcanBench Frontier v488.8%measured
AA AutomationBench74.0%estimated ± 4.8 pp, medium confidence
GDP.pdf27.0%estimated ± 2.8 pp, medium confidence
AA Terminal-Bench 4.060.5%estimated ± 6.2 pp, medium confidence
ARC-AGI-352.7%measured
ARC-AGI-3 Provider Adapter96.2%measured
cursorBench4144.8%estimated ± 3.7 pp, medium confidence
CWE-bench63.2%estimated ± 4.3 pp, medium confidence
MKA47.7%measured
SkillsBench official leaderboard (BenchFlow)59.5%estimated ± 6.1 pp, low confidence
BinaryAudit35.8%estimated ± 12.1 pp, low confidence
OTelBench22.5%estimated ± 6.6 pp, low confidence
FACTS Multimodal52.6%measured
Hermes Bench73.9%measured
Hermes Index57.1%measured
ParseBench-276.2%estimated ± 3.5 pp, high confidence
CharXiv87.3%estimated ± 5.6 pp, low confidence
CharXiv w/o tools84.6%estimated ± 1.8 pp, medium confidence
ExploitBench99.7%measured
ExploitGym35.1%measured
AI2D_TEST94.2%estimated ± 2.0 pp, low confidence
C-Eval93.4%estimated ± 1.5 pp, low confidence
CountBench100.0%estimated ± 2.9 pp, low confidence
DynaMath85.5%estimated ± 3.0 pp, low confidence
ERQA68.8%estimated ± 3.1 pp, medium confidence
HMMT Feb 2026100.0%estimated ± 3.1 pp, medium confidence
HMMT Nov 202596.2%estimated ± 1.5 pp, low confidence
MathVision95.0%estimated ± 3.1 pp, medium confidence
MCP Atlas82.1%estimated ± 10.7 pp, low confidence
MMAnswerBench97.7%estimated ± 2.4 pp, low confidence
MMLU-Redux98.7%estimated ± 0.8 pp, low confidence
MMMU-Pro85.3%estimated ± 1.5 pp, medium confidence
NL2Repo61.6%estimated ± 6.0 pp, medium confidence
PostTrainBench v142.5%estimated ± 1.3 pp, low confidence
RealWorldQA89.7%estimated ± 2.9 pp, low confidence
ScreenSpot Pro89.1%estimated ± 10.1 pp, low confidence
SuperGPQA100.0%estimated ± 9.3 pp, low confidence
Toolathlon55.9%estimated ± 6.3 pp, medium confidence
V*100.0%estimated ± 7.5 pp, low confidence
Video-MME (w/o subtitle)91.3%estimated ± 1.7 pp, low confidence
Video-MME (with subtitle)91.8%estimated ± 3.7 pp, low confidence
HealthBench Hard36.2%measured
MedXpertQA (MM)77.8%estimated ± 6.5 pp, low confidence
React Native Evals78.5%estimated ± 4.7 pp, medium confidence
ResearchClawBench21.6%estimated ± 2.2 pp, low confidence
PostTrainBench38.8%measured
DAYJOB15.5%measured
DAYJOB Healthcare10.8%measured
Finance Agent v262.3%estimated ± 3.8 pp, medium confidence
BioMysteryBench (human-solvable)89.5%estimated ± 0.6 pp, medium confidence
HealthBench (raw)56.7%measured
HealthBench (length-adjusted)58.5%measured
HealthBench Professional64.2%measured
HealthBench Professional (raw)67.2%measured
HLE-Verified55.4%estimated ± 1.3 pp, low confidence
LABBench283.1%estimated ± 1.7 pp, medium confidence
ScarfBench68.4%measured
BlenderWorld (text-only)85.3%measured
Cookie-Bench89.7%measured
STEMBench2.8%measured
Agents' Last Exam51.7%estimated ± 9.6 pp, medium confidence
BabyVision w/ Python94.4%estimated ± 4.0 pp, low confidence
OpenHarmony Bench v1.058.9%estimated ± 2.4 pp, high confidence
OpenHarmony Bench v1.263.8%estimated ± 11.0 pp, low confidence
SEC-Bench Pro78.8%measured
MRCRv285.9%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.5%estimated ± 2.6 pp, low confidence
SWE-Marathon59.1%estimated ± 2.1 pp, low confidence
BioBasicsBench62.6%measured
LifeVal80.1%estimated ± 2.3 pp, medium confidence
MMMU-Pro w/ Python88.1%estimated ± 2.0 pp, medium confidence
ExploitBench Internal Port21.5%measured
SWE-Atlas Codebase QnA61.0%estimated ± 4.8 pp, high confidence