LongCat-Flash-Omni

LongCat-Flash-Omni places in the top three on 2 benchmarks.

Meituan Released 31 Oct 2025 Family LongCat-Flash The lab's model card

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Meituan Oct 2025

Open weights Flagship

No Epoch Capabilities Index score yet

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	90.3%	#12 of 127	Model card
C-Eval	91.7%	#10 of 38	Model card
MMLU-Pro	82.7%	#48 of 151	Model card
DROP	83.5%	#16 of 40	Model card
CMMLU	89.4%	#7 of 17	Model card
IFEval	82.4%	#89 of 126	Model card
GraphWalks	56%	#19 of 21	Model card
COLLIE	45.7%	#11 of 12	Model card
Meeseeks	39%	#2 of 2	Model card

Benchmark	Score	Rank	Source
MATH-500	97.6%	#12 of 67	Model card
GPQA Diamond	74.4%	#144 of 284	Model card
ZebraLogic	86%	#12 of 23	Model card
AIME 2024	72.9%	#55 of 87	Model card
Beyond AIME	47.4%	#13 of 14	Model card

Benchmark	Score	Rank	Source
MBPP+	80.2%	#5 of 15	Model card
HumanEval+	90.8%	#10 of 27	Model card
LiveCodeBench	52.6%	#41 of 73	Model card
LiveCodeBench v6	52.6%	#68 of 76	Model card

Benchmark	Score	Rank	Source
AndroidControl	91.2%	#1 of 2	Model card
ScreenSpot-v2	91.2%	#10 of 14	Model card
VisualWebBench	78.7%	#2 of 2	Model card

Benchmark	Score	Rank	Source
RefCOCO / RefCOCO+ / RefCOCOg	93.9%	#2 of 19	Model card
MuirBench	77.1%	#4 of 12	Model card
ChartQA	87.6%	#14 of 37	Model card
MathVista	77.9%	#27 of 64	Model card
MMBench	87.5%	#4 of 9	Model card
MVBench	75.2%	#7 of 15	Model card
MMAU	75.9%	#7 of 14	Model card
MM-Vet	69%	#4 of 8	Model card
Mantis-Eval	84.8%	#1 of 2	Model card
OmniBench	61.4%	#1 of 2	Model card
OpenAudioBench	80.1%	#1 of 2	Model card
BLINK	63.1%	#15 of 28	Model card
TempCompass	82.2%	#4 of 7	Model card
MMMU	70.7%	#56 of 97	Model card
CountBench	92.4%	#12 of 20	Model card
VoiceBench	88.7%	#6 of 10	Model card
WorldSense (omnimodal)	60.9%	#3 of 5	Model card
RealWorldQA	74.8%	#19 of 30	Model card
Video-MME	78.2%	#28 of 42	Model card
Daily-Omni	82.4%	#4 of 6	Model card
Clotho-AQA	72.8%	#2 of 3	Model card
DocVQA	91.8%	#23 of 33	Model card
CoVoST2	37.3	#8 of 11	Model card
LongVideoBench	69.3%	#8 of 11	Model card
MMStar	70.9%	#22 of 30	Model card
OCRBench	84.9%	#21 of 28	Model card
AISHELL-1	0.63%	#3 of 4	Model card
MMVU	67.1%	#17 of 20	Model card
Video-MMMU	67.5%	#26 of 30	Model card
AISHELL-2	2.78%	#3 of 3	Model card
NExT-QA	86.2%	#1 of 1	Model card
Real-time Audio-Visual Interaction Evaluation	1.37	#1 of 1	Model card
UNO-Bench	49.9%	#1 of 1	Model card

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

Every result in the longer documents the lab published with it. 2 of them aren't in the launch post.