Gemini 3 Flash

Gemini 3 Flash ranks first on 6 of the 170 benchmarks it has scores on.

Google DeepMind Released 17 Dec 2025 Family Gemini 3 The lab's model card

172

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Google DeepMind Dec 2025

Gemini 3 Flash

Closed weights Mid-size

152.0 #54 of 271

Epoch Capabilities Index

SWE-bench Verified 75.4% #13 /32
SWE-Bench Pro 34.6% #15 /24
Terminal-Bench 2.0 64.3% #9 /43

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

Benchmark	Score	Rank	Source
MMLU-Pro	88.6%	#5 of 212	Official leaderboard
MMMLU	91.8%	#4 of 62	Model card
FACTS Search	81%	#3 of 31	Official leaderboard
LMArena (Elo)	1,473	#38 of 333	Official leaderboard
SimpleQA Verified	66.8%	#11 of 84	Epoch AI
Judgemark v2.1	85.4	#6 of 42	Official leaderboard
FACTS Benchmark Suite	64.9%	#6 of 40	Official leaderboard
LOCA-Bench	20%	#2 of 9	Official leaderboard
FACTS Parametric	72.3%	#11 of 45	Official leaderboard
Global PIQA	92.8%	#2 of 8	Model card
SkateBench v2	76.2%	#6 of 21	Official leaderboard
FACTS Parametric V2	72.4%	#7 of 20	Official leaderboard
Humanity's Last Exam	43.5%	#52 of 147	Model card
Extended NYT Connections	83.1%	#35 of 97	Official leaderboard
FACTS Multimodal	41.3%	#15 of 39	Official leaderboard
FACTS Search V2	80.9%	#11 of 21	Official leaderboard
MRCR v2	67.2%	#16 of 27	Model card
Longform Creative Writing (EQ-Bench)	55.7	#76 of 119	Official leaderboard
Judgemark v4	46.1	#28 of 40	Official leaderboard
Document Arena	1,439	#30 of 37	Official leaderboard
Vectara Hallucination Leaderboard (2025 dataset)	13.5%	#88 of 104	Official leaderboard
BullshitBench V2	10%	#108 of 122	Official leaderboard
BullshitBench	10.9%	#98 of 107	Official leaderboard

Math and science

Benchmark	Score	Rank	Source
BRUMO 2025	100%	#1 of 42	Official leaderboard
SMT 2025	92.9%	#2 of 40	Official leaderboard
AIME 2025	99.7%	#10 of 143	Model card
General365	60.8%	#2 of 23	Official leaderboard
HMMT 2025	97.5%	#5 of 53	Official leaderboard
AIME 2026	96.7%	#6 of 31	Official leaderboard
OTIS Mock AIME 2024-2025	95.6%	#40 of 202	Epoch AI
FrontierMath	35.6%	#14 of 70	Epoch AI
GPQA Diamond	89.4%	#48 of 219	Epoch AI
CMIMC 2025	90%	#9 of 33	Official leaderboard
DTBench	81.8%	#54 of 178	Official leaderboard
HMMT Nov 2025	93.3%	#7 of 22	Official leaderboard
SimpleBench	61.1%	#33 of 100	Official leaderboard
MathArena (Apex, IMO, USAMO)	15.6%	#16 of 47	Official leaderboard
HMMT 2026	89.4%	#14 of 31	Official leaderboard
ARC-AGI	84.7%	#43 of 92	Official leaderboard
ARC-AGI-2	33.6%	#47 of 95	Official leaderboard
FrontierMath Tiers 1-3 v2	51.2%	#53 of 89	Epoch AI
FrontierMath Tier 4	4.17%	#34 of 55	Epoch AI
Human Behavior Bench	78.5	#7 of 11	Official leaderboard
ProofBench	15%	#38 of 59	Official leaderboard
FrontierMath Tier 4 v2	17.1%	#51 of 71	Epoch AI
ForecastBench	64.6	#69 of 86	Official leaderboard
Verifying the Verifiers (Project APE CRED benchmark)	47%	#19 of 23	Official leaderboard

Coding

Benchmark	Score	Rank	Source
SWE-bench Multilingual	72.7%	#1 of 13	Official leaderboard
ALE-Bench	1,367	#28 of 121	Epoch AI
IOI (Vals)	39.1%	#13 of 55	Official leaderboard
WeirdML	61.6%	#29 of 112	Official leaderboard
OTelBench	18.8%	#4 of 14	Official leaderboard
CooperBench	27.8%	#2 of 6	Official leaderboard
SWE-fficiency	10.6%	#6 of 17	Official leaderboard
SWE-bench Verified	75.4%	#13 of 32	Epoch AI
ProgramBench	0%	#13 of 25	Official leaderboard
Code Arena	1,440	#66 of 124	Official leaderboard
GSO-Bench	9.80%	#17 of 30	Official leaderboard
OpenHands Index	49%	#20 of 33	Official leaderboard
SWE-Bench Pro	34.6%	#15 of 24	Official leaderboard
AndroidBench	62.5%	#22 of 32	Official leaderboard
WebDev Arena	1,383	#74 of 105	Epoch AI
Vibe Code Bench	20.2%	#82 of 115	Official leaderboard
LiveCodeBench Pro	2,316	#3 of 4	Model card
Image-to-WebDev Arena	1,450	#42 of 55	Official leaderboard
SWE Atlas Test Writing	30.3%	#17 of 22	Official leaderboard
Code Migration	6.39%	#69 of 78	Official leaderboard
SWE Atlas	8.20%	#22 of 22	Official leaderboard
SWE Atlas Refactoring	10%	#17 of 17	Official leaderboard

Agents and tools

Benchmark	Score	Rank	Source
PinchBench	87%	#1 of 48	Official leaderboard
VITA-Bench	32.5%	#1 of 18	Official leaderboard
GameWorld	41.9%	#1 of 12	Official leaderboard
Elimination Game	6	#5 of 56	Official leaderboard
Game Arena Chess Openings	1,263	#1 of 11	Official leaderboard
Game Arena Coin Game	472	#4 of 35	Official leaderboard
SnakeBench	32	#29 of 183	Official leaderboard
Toolathlon	49.4%	#8 of 50	Official leaderboard
Agent Poker Bench	1,100	#3 of 17	Official leaderboard
Game Arena Dark Hex	348	#7 of 38	Official leaderboard
BalatroBench	16.7	#3 of 16	Official leaderboard
BALROG	48.1%	#7 of 35	Official leaderboard
Terminal-Bench 2.0	64.3%	#9 of 43	Official leaderboard
tau2-bench	90.2%	#15 of 69	Model card
SkillsBench	54.6%	#5 of 21	Official leaderboard
Game Arena Chess	1,212	#10 of 39	Official leaderboard
DeepResearch Bench	49.8%	#7 of 24	Epoch AI
LisanBench	1996.0	#43 of 133	Official leaderboard
Game Arena Word Art	678	#12 of 35	Official leaderboard
Game Arena Five in a Row	756	#7 of 20	Official leaderboard
GAIA	68.4%	#20 of 57	Official leaderboard
Game Arena Werewolf	-1	#3 of 8	Official leaderboard
Pencil Puzzle Bench	6.70%	#21 of 53	Official leaderboard
Chess Puzzles	39%	#36 of 83	Epoch AI
Mystery Game Puzzles	26%	#35 of 80	Epoch AI
Game Arena Word Association	259	#18 of 36	Official leaderboard
CUA-World-Long	14%	#2 of 4	Official leaderboard
ScreenSpot-Pro	69.1%	#13 of 25	Model card
PinchBench 2.0	72.1%	#30 of 57	Official leaderboard
Game Arena Bargaining	186	#19 of 36	Official leaderboard
Kaggle Game Arena	278	#21 of 39	Official leaderboard
Game Arena Checkers	892	#21 of 38	Official leaderboard
Game Arena Nine Men's Morris	582	#20 of 36	Official leaderboard
τ³-bench Banking	27.3%	#14 of 25	Official leaderboard
tau3-bench	83.5%	#4 of 7	Official leaderboard
Game Arena Reversi	336	#23 of 36	Official leaderboard
Game Arena Poker	-3.86	#20 of 31	Official leaderboard
Game Arena Four in a Row	256	#26 of 38	Official leaderboard
UniClawBench	28.3%	#7 of 10	Official leaderboard
Game Arena Dots and Boxes	757	#26 of 37	Official leaderboard
LMArena Search Arena	1,189	#25 of 35	Official leaderboard
LLM Chess	-179	#92 of 124	Official leaderboard
Game Arena Go	497	#28 of 36	Official leaderboard
Game Arena Ultimate Tic-Tac-Toe	169	#28 of 36	Official leaderboard
MCP Atlas	62%	#27 of 34	Official leaderboard
Claw-Eval Pass^3	49.2%	#19 of 23	Official leaderboard
CEO-Bench	$0	#16 of 19	Official leaderboard
Claw-Eval	72.3%	#21 of 23	Official leaderboard
Game Arena Lines of Action	277	#33 of 36	Official leaderboard
Game Arena Clobber	21	#35 of 36	Official leaderboard

Professional work

Benchmark	Score	Rank	Source
MedCode	55.9%	#5 of 96	Official leaderboard
LegalBench	86.9%	#11 of 138	Official leaderboard
MortgageTax	68.7%	#11 of 86	Official leaderboard
CorpFin v2	66.4%	#20 of 119	Official leaderboard
ProfBench	53.4%	#21 of 72	Official leaderboard
TaxEval v2	73.9%	#44 of 129	Official leaderboard
Vals Finance Agent v1	39.3%	#19 of 47	Official leaderboard
prinzbench	36.0	#13 of 26	Official leaderboard
CaseLaw v2	55.8%	#32 of 53	Official leaderboard
Vending-Bench Arena	$3,436	#20 of 32	Official leaderboard
Vals Finance Agent	47.6%	#31 of 49	Official leaderboard
PolicyBench	80.6%	#32 of 50	Official leaderboard
Vending-Bench 2	$3,635	#44 of 66	Official leaderboard
Legal Research Bench	18.3%	#65 of 88	Official leaderboard
Vals Finance Agent v2	42.5%	#60 of 80	Official leaderboard
Tax Agent Bench	11.8%	#61 of 79	Official leaderboard
Harvey LAB (Legal Agent Benchmark)	0%	#64 of 80	Official leaderboard
MedScribe	69.9%	#86 of 98	Official leaderboard
Excel Modeling Benchmark	26%	#68 of 76	Official leaderboard
EEBench	19.5%	#24 of 26	Official leaderboard
CommerceAgentBench	29%	#13 of 13	Official leaderboard

Vision, audio and robotics

Benchmark	Score	Rank	Source
MATH-Vision (MathVision)	90.6%	#4 of 105	Official leaderboard
GeoBench	88%	#2 of 48	Official leaderboard
ParseBench	75%	#3 of 42	Official leaderboard
VideoMME v2 (with subtitles)	42.5%	#3 of 31	Official leaderboard
VPCT	72.6%	#3 of 28	Epoch AI
MMMU-Pro	81.2%	#8 of 72	Model card
Video-MMMU	86.9%	#4 of 30	Model card
OCR Arena	1,770	#2 of 11	Official leaderboard
Vision Arena	1,273	#28 of 136	Official leaderboard
ZeroBench	15%	#19 of 66	Official leaderboard
1H-VideoQA	79.2%	#6 of 20	Official leaderboard
Math Kangaroo 2025 (MathArena)	85.8%	#8 of 22	Official leaderboard
OmniDocBench v1.6	92.6	#3 of 7	Official leaderboard
Claw-Eval-MM	50.4%	#5 of 11	Official leaderboard
RoboRewardBench	0.92	#9 of 19	Official leaderboard
CharXiv	80.3%	#34 of 67	Model card
OmniDocBench v1.5	0.12	#10 of 13	Model card
Blueprint-Bench 2	0%	#32 of 35	Official leaderboard
GDP.pdf	10%	#36 of 39	Epoch AI
VoxelBench	1,275	#52 of 56	Official leaderboard

Safety and security

Benchmark	Score	Rank	Source
BinaryAudit	36.8%	#7 of 20	Official leaderboard
Human Pathogen Capabilities Test v2 (HPCT v2)	56%	#33 of 71	Official leaderboard
CRSBench	56.9%	#3 of 5	Official leaderboard
KORA	24%	#32 of 42	Official leaderboard

Other

Benchmark	Score	Rank	Source
SAGE (Vals)	51.9%	#10 of 81	Official leaderboard
Conceptual Reasoning Index	56.0	#47 of 149	Official leaderboard
LMCA	43.1%	#45 of 139	Epoch AI
CAIS Vision Capabilities Index	59.3	#18 of 53	Official leaderboard
ACCoRD	68.6	#55 of 150	Official leaderboard
CAIS Text Capabilities Index	36.5	#27 of 66	Official leaderboard
Vals Multimodal Index	52.2%	#24 of 33	Official leaderboard
CAIS Risk Index	66.3	#59 of 65	Official leaderboard

Google DeepMind's models over time

Google DeepMind's models in color, against every other model on the Epoch Capabilities Index, by release date.

Its launch post

The benchmarks the lab chose to headline, with the scores it gave.

Gemini 3 Flash model card 17 Dec 2025