Ling-2.6-1T

Ling-2.6-1T.

Ant Group Released April 29th, 2026 Family Ling 2.6 The lab's model card Compare with others

Benchmarks with a score

Ranked first, out of five or more models

In the top three

Ant Group Apr 2026

Ling-2.6-1T

Open weights Flagship

140.2 #176 of 404

Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.

IFBench 57.6% #58 /80
BFCL V4 70.6% #8 /30
LongBench v2 48.3% #28 /35

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
MRCR v2	80.4%	#9 of 30	Lab model card
Chinese SimpleQA	76.5%	#5 of 13	Lab model card
BBEH micro average	52.4%	#6 of 15	Lab model card
SuperGPQA	58.3%	#20 of 35	Lab model card
IFBench	57.6%	#58 of 80	Lab model card
SimpleQA Verified	31.5%	#17 of 22	Lab model card
LongBench v2	48.3%	#28 of 35	Lab model card
Humanity's Last Exam	10.1%	#142 of 154	Lab model card
Extended NYT Connections	4.10%	#103 of 103	Official leaderboard

Math and science

Benchmark	Score	Rank	Source
GPQA Diamond	76.2%	#142 of 302	Lab model card
AIME 2026	87.4%	#30 of 38	Lab model card
HMMT Nov 2025	81.9%	#23 of 28	Lab model card
ARC-AGI	50.9%	#19 of 21	Lab model card
IMO-AnswerBench	65.8%	#35 of 37	Lab model card

Coding

Benchmark	Score	Rank	Source
SWE-bench Verified	72.2%	#64 of 155	Lab model card
ALE-Bench	1,108	#93 of 126	Official leaderboard
LiveCodeBench v6	65.6%	#68 of 90	Lab model card

Agents and tools

Benchmark	Score	Rank	Source
PinchBench 2.0	82.6%	#13 of 57	Official leaderboard
BFCL V4	70.6%	#8 of 30	Lab model card
SnakeBench	25	#86 of 188	Official leaderboard
tau2-bench	78.4%	#45 of 79	Lab model card
PinchBench	87.4%	#8 of 14	Lab model card
ClawProBench	57.4%	#32 of 49	Official leaderboard
Terminal-Bench 2.0	40.5%	#52 of 63	Lab model card
Claw-Eval Pass^3	51%	#12 of 13	Lab model card

Safety and security

Benchmark	Score		Rank	Source
nocot-bench (NCRI)	73.9		#81 of 185	Official leaderboard

Its lab's numbers put it at 140.2, number 176 of 404

The Self-Reported Index fits one score to the 20 capability results Ant Group reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 137.9 to 142.6. How we compute it.

Its launch post

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

Ling-2.6-1T model card (Hugging Face) Apr 29, 2026

Its system card

Every result in the longer documents the lab published with it. 12 of them aren't in the launch post.

Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale Jun 13, 2026