Ling-2.6-flash is an open-source instruct large language model release from inclusionAI/Ant Group, using a hybrid-linear sparse MoE architecture with 104B total parameters and 7.4B active parameters, optimized for fast, token-efficient agent workflows, coding, document processing, tool use, and long-context language tasks.
Pricing
Model profile · Current snapshot
Input / 1M
$0.10
Output / 1M
$0.30
Blended / 1M
$0.15
Output TPS
84.14
TTFT (s)
0.88
Model Intelligence
Context window
262,144 tokens
Arena ranking
14
Benchmarkable
Yes
Model level
release
Intelligence Index
14.2
Coding Index
25.3
GPQA
0.59
HLE
0.06
SciCode
0.27
IFBench
0.57
LCR
0.28
TerminalBench Hard
0.21
TAU2
0.86
Recent stories
1 linked story