Model

GPT-5.6 Sol

OpenAI · closed weights

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
AI Consilium	2.8	mean_judge_score	structured	Independent	medium	arxiv.org
Balrog	60.0	average_progress	N/A	Independent	medium	balrogai.com
CritPt	32.3	accuracy	N/A	Independent	low	epoch.ai
EnigmaEval	37.1	accuracy	N/A	Independent	medium	labs.scale.com
GPQA Diamond	93.5	accuracy	N/A	Independent	high	epoch.ai
HouseholdBench	0.9	relmae	N/A	Independent	medium	arxiv.org
InsClaimBench	80.0	accuracy	N/A	Independent	medium	arxiv.org
LexPilot	0.7	lca_macro-f1	N/A	Independent	medium	arxiv.org
MechHypoBench	64.9	mca_(%)	codex	Independent	medium	arxiv.org
MOF-VERIFY	62.8	macro-f1	N/A	Independent	medium	arxiv.org
ScholarCatalyst	0.5	r@20	tool-calling agent	Independent	medium	arxiv.org
SDDBench	92.4	accuracy	N/A	Independent	medium	arxiv.org
SEMAADB	80.7	f1	N/A	Independent	medium	arxiv.org
SimpleQA Verified	69.7	accuracy	N/A	Independent	high	epoch.ai
TasteVal	0.6	compute_multiplier	N/A	Independent	medium	arxiv.org

Math

Benchmark	Score	Metric	Harness	Source	Confidence	Link
Australian SOC Benchmark	84.0	t2_accuracy	N/A	Independent	medium	arxiv.org
GIFT-Eval-CTX	0.6	mase	N/A	Independent	medium	arxiv.org
OpenProblemBench	8.5	mean_judged_solve_rate	N/A	Independent	medium	arxiv.org

Coding

Benchmark	Score	Metric	Harness	Source	Confidence	Link
A2Z GameSpec-Bench	61.8	gdd_fidelity	N/A	Independent	medium	arxiv.org
AgentSWE	13.7	score	N/A	Independent	medium	arxiv.org
ArchitectureIQ	44.0	accuracy	architectureiq	Independent	medium	arxiv.org
ArrivalBench	7.0	failure_rate	N/A	Independent	medium	arxiv.org
CRBugBench	43.5	success_rate	N/A	Independent	medium	arxiv.org
CRUXEval-style	74.8	accuracy	N/A	Independent	medium	arxiv.org
D2K-Bench	2.0	s_perf	N/A	Independent	medium	arxiv.org
DeepSWE	72.7	pass@1	mini-swe-agent	Independent	medium	deepswe.datacurve.ai
Doc2Harness	43.8	pass_rate	N/A	Independent	medium	arxiv.org
E2E-SWE	61.2	pass@1	N/A	Independent	medium	arxiv.org
GSO	76.5	score_opt@1	OpenHands	Independent	medium	gso-bench.github.io
MathVis	21.5	pass@1	N/A	Independent	medium	arxiv.org
MetaKernelBench	56.8	paired_lift	cute dsl	Independent	medium	arxiv.org
ModeBench	0.2	pcmd	N/A	Independent	medium	arxiv.org
SciCode	57.1	score	N/A	Independent	medium	artificialanalysis.ai
SpecGuardBench	81.8	accuracy	codex	Independent	medium	arxiv.org
SWEPatchTest	64.1	f1	N/A	Independent	medium	arxiv.org
TestSuiteBench	37.7	score	N/A	Independent	medium	arxiv.org
Video2World	31.0	v2wscore	N/A	Independent	medium	arxiv.org
WorldSolver	48.7	score	N/A	Independent	medium	arxiv.org

Agentic / Tool Use

Benchmark	Score	Metric	Harness	Source	Confidence	Link
AgenticBBO-Bench	0.6	hpo	N/A	Independent	medium	arxiv.org
AgentMonBench	1.0	silentswap	N/A	Independent	medium	arxiv.org
AgentPersonaBench	74.1	accuracy	N/A	Independent	medium	arxiv.org
APTInvestBench	71.8	coverage_rate	N/A	Independent	medium	arxiv.org
AutoDataBench	58.9	performance_of_models_trained_on_optimized_data_vs_baseline_prediction_accuracy_comparison	N/A	Independent	medium	arxiv.org
BOTTLED	0.1	f1	N/A	Independent	medium	arxiv.org
BrickBench	1015.0	elo	N/A	Independent	medium	arxiv.org
CineRecon	39.7	score	codex cli 0.153.4	Independent	medium	arxiv.org
CompMat-Bench	90.4	pass_rate	matclaw	Independent	medium	arxiv.org
cua-speedrun	81.6	success_rate	N/A	Independent	medium	arxiv.org
DelegationBench	58.1	accuracy	N/A	Independent	medium	arxiv.org
DrivingBench	6.0	progress	codex	Independent	medium	arxiv.org
DrugTargetWorld	15.8	score	N/A	Independent	medium	arxiv.org
DUDA-Bench	0.1	workflow_score	react	Independent	medium	arxiv.org
DyadMem	70.3	accuracy	N/A	Independent	medium	arxiv.org
EngiWorld	38.0	engiscore	N/A	Independent	medium	arxiv.org
EurekaBench	0.1	tau_b	N/A	Independent	medium	arxiv.org
GraceBench	0.4	recovery_ability	N/A	Independent	medium	arxiv.org
HICO-Bench	76.3	success_rate	N/A	Independent	medium	arxiv.org
HygieneRoboBench	78.5	resolved_rate	N/A	Independent	medium	arxiv.org
HyperBrowseComp	19.1	accuracy	N/A	Independent	medium	arxiv.org
IGMBench	78.2	success_rate	N/A	Independent	medium	arxiv.org
KlinikeBench	23.7	strict_pass@1	N/A	Independent	medium	arxiv.org
Learn2Play Bench	59.5	max	codex	Independent	medium	arxiv.org
Learn2Play Bench	46.1	max	opencode	Independent	medium	arxiv.org
LifeSearch	63.4	score	N/A	Independent	medium	arxiv.org
Mine Odyssey	43.9	success_rate	N/A	Independent	medium	arxiv.org
MisVisAgentBench	49.3	pra	N/A	Independent	medium	arxiv.org
MMPostTrainBench	44.4	mean_task_score_(%)	codex	Independent	medium	arxiv.org
MS-Eval	38.0	avg.	mad-sae (spec)	Independent	medium	arxiv.org
OfficeTown Bench	33.5	pass@1	N/A	Independent	medium	arxiv.org
OpenGameEval	47.3	pass_rate	N/A	Independent	medium	arxiv.org
PhysClosureBench	5.8	mean_score	N/A	Independent	medium	arxiv.org
PhysCodeGym	0.6	s_tot	codex	Independent	medium	arxiv.org
PlanBench	65.8	accuracy	N/A	Independent	medium	arxiv.org
RealCLI	35.7	score	N/A	Independent	medium	arxiv.org
Realtime-Bench	18.0	success_rate	N/A	Independent	medium	arxiv.org
StoreBench	0.5	composite	N/A	Independent	medium	arxiv.org
TiGanEval	64.1	success_rate	N/A	Independent	medium	arxiv.org
TrialPD-Bench	80.0	f1	N/A	Independent	medium	arxiv.org

Multimodal

Benchmark	Score	Metric	Harness	Source	Confidence	Link
3DVerifierBench	64.9	accuracy	N/A	Independent	medium	arxiv.org
ChartBmkAgent	84.3	accuracy	N/A	Independent	medium	arxiv.org
ChemArgus	29.8	full_problem	text only	Independent	medium	arxiv.org
CREATE	27.7	accuracy	N/A	Independent	medium	arxiv.org
EgoProbe	55.3	accuracy	N/A	Independent	medium	arxiv.org
Event-STVG	2.3	m-viou	N/A	Independent	medium	arxiv.org
EviLens	51.3	accuracy	N/A	Independent	medium	arxiv.org
FastBench	46.7	accuracy	N/A	Independent	medium	arxiv.org
Game2Real-Bench	69.8	accuracy	N/A	Independent	medium	arxiv.org
GDP.pdf	30.7	gdp.pdf_score	N/A	Independent	medium	surgehq.ai
GeoSpatialBench	78.1	accuracy	N/A	Independent	medium	arxiv.org
HeiCo-FOCUS	46.8	accuracy	N/A	Independent	medium	arxiv.org
NextAct	0.3	embedding_score	N/A	Independent	medium	arxiv.org
OmniCompass	44.5	cs_avg	N/A	Independent	medium	arxiv.org
OmniVCBench	55.1	accuracy	N/A	Independent	medium	arxiv.org
PlotGround-1k	73.8	accuracy	N/A	Independent	medium	arxiv.org
RoTA-Bench	17.0	success_rate	N/A	Independent	medium	arxiv.org
SEAR	89.0	success_rate	N/A	Independent	medium	arxiv.org

Long Context

Benchmark	Score	Metric	Harness	Source	Confidence	Link
LongHarness Bench	60.0	accuracy	direct	Independent	medium	arxiv.org

Instruction Following

Benchmark	Score	Metric	Harness	Source	Confidence	Link
Design Creativity Bench	51.3	overall	N/A	Independent	medium	arxiv.org
Nonlinear Narrative Generation Benchmark	0.7	acc.	N/A	Independent	medium	arxiv.org

Safety & Alignment

Benchmark	Score	Metric	Harness	Source	Confidence	Link
APEX-Harm	20.0	success_rate	N/A	Independent	medium	arxiv.org
DecepEval	36.8	deception_rate	N/A	Independent	medium	arxiv.org
DistressBench	0.7	score	N/A	Independent	medium	arxiv.org
DRYLAB-Bench	0.3	floor-included_brs_s2	N/A	Independent	medium	arxiv.org
EvoRiskBench	69.8	success_rate	N/A	Independent	medium	arxiv.org
FIGS	29.1	automated_judge_applying_a_10-rule_taxonomy_separating_sycophancy_from_calibrated_validation	N/A	Independent	medium	arxiv.org
Guise	0.7	unsafe_compliance_rate	N/A	Independent	medium	arxiv.org
Nonpublic Vulnerabilities Benchmark	0.1	attack_score	codex	Independent	medium	arxiv.org
ObligationBench	47.8	recall	N/A	Independent	medium	arxiv.org

Abstract Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
ARC-AGI-2	92.5	score	N/A	Independent	low	epoch.ai
Grid-Bench	81.6	accuracy	N/A	Independent	medium	arxiv.org

Multilingual

Benchmark	Score	Metric	Harness	Source	Confidence	Link
CSWBench	17.8	cswbench_extended_/20	N/A	Independent	medium	arxiv.org
EHR-Bench	43.8	jdg_score	N/A	Independent	medium	arxiv.org
HakemBench	0.8	composite	N/A	Independent	medium	arxiv.org

Knowledge & Reasoning

Math

Coding

Agentic / Tool Use

Multimodal

Long Context

Instruction Following

Safety & Alignment

Abstract Reasoning

Multilingual

Reproduced by the community