Model

DeepSeek V4.1 Flash

DeepSeek · open weights

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
CritPt	14.3	accuracy	N/A	Independent	low	epoch.ai
GPQA Diamond	89.8	accuracy	N/A	Independent	high	epoch.ai
InsClaimBench	77.5	accuracy	N/A	Independent	medium	arxiv.org
SDDBench	92.8	accuracy	N/A	Independent	medium	arxiv.org
SimpleQA Verified	42.9	accuracy	N/A	Independent	high	epoch.ai

Benchmark	Score	Metric	Harness	Source	Confidence	Link
OpenProblemBench	1.2	mean_judged_solve_rate	N/A	Independent	medium	arxiv.org
SELF-POT	0.8	success_rate	N/A	Independent	medium	arxiv.org

Benchmark	Score	Metric	Harness	Source	Confidence	Link
DesignReviewBench	0.3	f1	designlore	Independent	medium	arxiv.org
Doc2Harness	40.2	pass_rate	N/A	Independent	medium	arxiv.org
PremiseSQL	0.3	ex	N/A	Independent	medium	arxiv.org
SciCode	51.9	score	N/A	Independent	medium	artificialanalysis.ai
Video2World	30.5	v2wscore	N/A	Independent	medium	arxiv.org
WorldSolver	24.6	score	N/A	Independent	medium	arxiv.org

Benchmark	Score	Metric	Harness	Source	Confidence	Link
AgenticBBO-Bench	0.6	hpo	N/A	Independent	medium	arxiv.org
AgentMonBench	0.8	silentswap	N/A	Independent	medium	arxiv.org
Argo-Bench	25.4	score	N/A	Independent	medium	arxiv.org
BioStudyBench	76.0	pass_rate	N/A	Independent	medium	arxiv.org
BrickBench	1006.0	elo	N/A	Independent	medium	arxiv.org
CineRecon	53.9	score	dsh 0.1.5-rc.3	Independent	medium	arxiv.org
CompMat-Bench	66.0	pass_rate	matclaw	Independent	medium	arxiv.org
EngiWorld	25.5	engiscore	N/A	Independent	medium	arxiv.org
FDE-Gym	0.2	score	N/A	Independent	medium	arxiv.org
GraceBench	0.6	recovery_ability	N/A	Independent	medium	arxiv.org
HygieneRoboBench	90.7	resolved_rate	N/A	Independent	medium	arxiv.org
LIVIN	11.4	success_rate	N/A	Independent	medium	arxiv.org
Mine Odyssey	23.9	success_rate	N/A	Independent	medium	arxiv.org
PersonaCUABench	14.1	success_rate	N/A	Independent	medium	arxiv.org
PhysClosureBench	4.5	mean_score	N/A	Independent	medium	arxiv.org
PhysCodeGym	0.5	s_tot	claude code	Independent	medium	arxiv.org
RealCLI	45.6	score	N/A	Independent	medium	arxiv.org
Realtime-Bench	8.7	success_rate	N/A	Independent	medium	arxiv.org
RobotWorld	1.2	success_rate	N/A	Independent	medium	arxiv.org
RSE-Bench	95.4	score	alfworld ace	Independent	medium	arxiv.org
RSE-Bench	95.4	score	alfworld rethinkskill	Independent	medium	arxiv.org
RSE-Bench	95.4	score	alfworld skillopt	Independent	medium	arxiv.org
RSE-Bench	43.3	score	skillflow ace	Independent	medium	arxiv.org
ServeLearnBench	33.2	accuracy	skillopt	Independent	medium	arxiv.org
TiGanEval	81.3	success_rate	N/A	Independent	medium	arxiv.org

Benchmark	Score	Metric	Harness	Source	Confidence	Link
GDP.pdf	19.8	gdp.pdf_score	N/A	Independent	medium	surgehq.ai
GeoSpatialBench	49.5	accuracy	N/A	Independent	medium	arxiv.org
PlotGround-1k	29.3	accuracy	N/A	Independent	medium	arxiv.org

Benchmark	Score	Metric	Harness	Source	Confidence	Link
Design Creativity Bench	43.6	overall	N/A	Independent	medium	arxiv.org

Benchmark	Score	Metric	Harness	Source	Confidence	Link
Bemula	28.5	score	full context	Independent	medium	arxiv.org
CWE-bench	55.0	pass@1	opencode	Independent	medium	arxiv.org
FIGS	37.3	automated_judge_applying_a_10-rule_taxonomy_separating_sycophancy_from_calibrated_validation	N/A	Independent	medium	arxiv.org
ObligationBench	32.1	recall	N/A	Independent	medium	arxiv.org

Benchmark	Score	Metric	Harness	Source	Confidence	Link
ARC-AGI-2	72.9	score	N/A	Independent	low	epoch.ai
ArgGYM	89.7	accuracy	N/A	Independent	medium	arxiv.org