DeepSeek-V3.2

DeepSeek-V3.2.

DeepSeek Released 1 Dec 2025 Family DeepSeek-V3 The lab's model card

101

Benchmarks with a score

Ranked first, out of five or more models

In the top three

DeepSeek Dec 2025

Open weights Flagship

146.1 #85 of 271

Epoch Capabilities Index

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.

Benchmark	Score	Rank	Source
MMLU-Pro	85%	#28 of 151	Model card
Vectara Hallucination Leaderboard (2025 dataset)	6.30%	#27 of 104	Official leaderboard
LMArena (Elo)	1,425	#100 of 333	Official leaderboard
Creative Writing v3	71.7	#45 of 120	Official leaderboard
Humanity's Last Exam	40.8%	#59 of 147	Model card
Longform Creative Writing (EQ-Bench)	66.8	#50 of 119	Official leaderboard
LOCA-Bench	6.70%	#4 of 9	Official leaderboard
Judgemark v2.1	66.3	#25 of 42	Official leaderboard
SkateBench v2	46.7%	#15 of 21	Official leaderboard
Extended NYT Connections	36.7%	#73 of 97	Official leaderboard
BullshitBench	14.6%	#86 of 107	Official leaderboard
CL-bench Life	9.50%	#13 of 16	Official leaderboard
BullshitBench V2	13%	#101 of 122	Official leaderboard
CL-bench	13.2%	#22 of 24	Official leaderboard

Benchmark	Score	Rank	Source
HiPhO	65.7%	#5 of 35	Official leaderboard
BRUMO 2025	96.7%	#9 of 42	Official leaderboard
AIME 2025	93.1%	#32 of 143	Model card
OTIS Mock AIME 2024-2025	87.8%	#67 of 202	Epoch AI
GPQA Diamond	83.4%	#77 of 219	Epoch AI
DTBench	76%	#67 of 178	Official leaderboard
FrontierMath	22.1%	#27 of 70	Epoch AI
HMMT 2025	92.5%	#19 of 49	Model card
SMT 2025	87.7%	#16 of 40	Official leaderboard
General365	54.9%	#11 of 23	Official leaderboard
CMIMC 2025	84.4%	#17 of 33	Official leaderboard
HMMT Nov 2025	90%	#12 of 22	Official leaderboard
ARC-AGI	57%	#61 of 92	Official leaderboard
FrontierMath Tier 4	2.10%	#37 of 55	Epoch AI
AIME 2026	94.2%	#21 of 31	Official leaderboard
MathArena (Apex, IMO, USAMO)	2.08%	#32 of 47	Official leaderboard
ARC-AGI-2	4.03%	#71 of 95	Official leaderboard
IMO-AnswerBench	78.3%	#29 of 37	Model card
ProofBench	8.00%	#47 of 59	Official leaderboard
HMMT 2026	84.1%	#25 of 31	Official leaderboard
Leipzig Benchmark	8.00%	#17 of 19	Official leaderboard
Echo General AI Prediction Leaderboard	964	#16 of 17	Official leaderboard

Benchmark	Score	Rank	Source
LiveCodeBench	83.3%	#6 of 73	Model card
Aider Polyglot	74.2%	#8 of 54	Epoch AI
SWE-bench Verified	70%	#15 of 60	Official leaderboard
Codeforces rating	2,386	#12 of 41	Model card
StructEval	75.5%	#14 of 28	Official leaderboard
IOI (Vals)	14.4%	#30 of 55	Official leaderboard
SWE-bench Multilingual	70.2%	#42 of 70	Model card
OpenHands Index	45.7%	#21 of 33	Official leaderboard
FrontierCS	24.8	#10 of 14	Official leaderboard
OTelBench	11.6%	#10 of 14	Official leaderboard
Code Arena	1,363	#92 of 124	Official leaderboard
OJBench	51.7%	#6 of 8	Official leaderboard
SWE-Bench Pro	15.6%	#20 of 24	Official leaderboard
WebDev Arena	1,325	#89 of 105	Epoch AI
Vibe Code Bench	5.11%	#101 of 115	Official leaderboard

Benchmark	Score	Rank	Source
Agent Poker Bench	1,090	#4 of 17	Official leaderboard
LisanBench	2348.3	#40 of 133	Official leaderboard
VITA-Bench	24%	#7 of 18	Official leaderboard
BalatroBench	8.67	#7 of 16	Official leaderboard
LLM Chess	189	#55 of 124	Official leaderboard
tau2-bench	80.3%	#31 of 69	Model card
BrowseComp-ZH	65%	#9 of 17	Model card
Terminal-Bench 2.0	39.5%	#24 of 43	Official leaderboard
Toolathlon	35.2%	#29 of 50	Official leaderboard
BrowseComp	67.6%	#46 of 77	Model card
GAIA	37.2%	#36 of 57	Official leaderboard
SnakeBench	20	#129 of 183	Official leaderboard
RWSearch	74%	#5 of 7	Official leaderboard
Claw-Eval-Live	51.4%	#9 of 12	Official leaderboard
Pencil Puzzle Bench	0%	#40 of 53	Official leaderboard
PinchBench	59.9%	#39 of 48	Official leaderboard
Game Arena Poker	-16.86	#26 of 31	Official leaderboard
MCPMark	38%	#9 of 10	Model card
Game Arena Chess	84	#37 of 39	Official leaderboard
Game Arena Five in a Row	126	#19 of 20	Official leaderboard
Claw-Eval	67.5%	#22 of 23	Official leaderboard
Claw-Eval Pass^3	40.2%	#22 of 23	Official leaderboard
Game Arena Word Association	15	#35 of 36	Official leaderboard
Game Arena Checkers	59	#37 of 38	Official leaderboard
Game Arena Four in a Row	9	#37 of 38	Official leaderboard
Kaggle Game Arena	0	#39 of 39	Official leaderboard
Game Arena Dark Hex	0	#38 of 38	Official leaderboard
Game Arena Dots and Boxes	0	#37 of 37	Official leaderboard
Game Arena Chess Openings	0	#11 of 11	Official leaderboard
MCP-Universe	45.9%	#1 of 1	Model card

Benchmark	Score	Rank	Source
ProfBench	50.5%	#28 of 72	Official leaderboard
SpreadsheetBench 2	15.6%	#5 of 8	Official leaderboard
CaseLaw v2	55.4%	#34 of 53	Official leaderboard
TaxEval v2	68.2%	#93 of 129	Official leaderboard
LegalBench	76.1%	#103 of 138	Official leaderboard
CorpFin v2	51%	#94 of 119	Official leaderboard
Vending-Bench 2	$1,034	#53 of 66	Official leaderboard
Vals Finance Agent	2.35%	#48 of 49	Official leaderboard
APEX-Agents	21.3%	#55 of 56	Epoch AI

Benchmark	Score	Rank	Source
KORA	24.3%	#31 of 42	Official leaderboard
BinaryAudit	12.3%	#16 of 20	Official leaderboard
Human Pathogen Capabilities Test v2 (HPCT v2)	37.4%	#58 of 71	Official leaderboard
SMDD-Bench	24.3%	#6 of 7	Official leaderboard
Agentic BAIM-LLM Evaluation (ABLE)	23.8%	#14 of 16	Official leaderboard
Dictatorship Eval	0.97%	#19 of 20	Official leaderboard

Benchmark	Score	Rank	Source
ACCoRD	64.6	#73 of 150	Official leaderboard
Conceptual Reasoning Index	45.4	#85 of 149	Official leaderboard
LMCA	28.8%	#88 of 139	Epoch AI
CAIS Text Capabilities Index	21.5	#48 of 66	Official leaderboard
CAIS Risk Index	68.3	#62 of 65	Official leaderboard

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

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

Every result in the longer documents the lab published with it.