LongCat Flash Lite
LongCatOpen WeightMIT · Commercial OK
Description
LongCat-Flash-Lite is a lightweight MoE model from Meituan with 68.5B total parameters and only 2.9B-4.5B activated per token. It explores N-gram embedding expansion as a new scaling direction, supporting 256K context length via YaRN. Optimized for agent tooling and programming tasks, achieving 500-700 tokens per second inference speed while maintaining strong performance on coding, math, and agentic benchmarks.
Release Date
2026-01-28
Parameters
68.5B
Context Length
—
Modalities
text
Capability Radar
70
general
40
coding
80
reasoning
60
scienceest.
70
agents
0
multimodal
Science is estimated from LLM Stats science scores or reasoning when no dedicated science benchmarks are available.
Rankings
No ranking data available
Benchmark Scores (LLM Stats)
(LLM Stats (zeroeval))Chat
Tau2 Airline
58.0%SR
Communication
Tau2 Retail
73.1%SR
Tau2 Telecom
72.8%SR
General
MMLU
85.5%SR
Language
CMMLU
82.5%SR
MMLU-Pro
78.3%SR
Math
MATH-500
96.8%SR
AIME 2024
72.2%SR
AIME 2025
63.2%SR
Reasoning
GPQANYU + Cohere + Anthropic (2023)
66.8%SR
SWE-Bench Verified
54.4%SR
SWE-bench Multilingual
38.1%SR
Terminal-Bench
33.8%SR
AA Evaluation Indices
(Artificial Analysis)No AA evaluation data available
LLM Stats Category Scores
(LLM Stats (zeroeval))Language80
Legal80
Math80
Finance80
Healthcare80
Physics70
Reasoning70
General70
Biology70
Chemistry70
Communication70
Tool Calling70
Chat60
Frontend Development50
Code40
Agents30
Pricing
Input Price$0.1 / 1M tokens
Output Price$0.4 / 1M tokens
Blended Price (3:1)$0.175 / 1M tokens
Speed
No speed data available
Provider Price Ranking
Provider Price Ranking
2 providers
Cheapest: LongCatMost Expensive: Meituan
ProviderInputOutput
1LongCatPRIMARY
$0.1
$0.4
2Meituan
$0.1
$0.4
Compare pricing across different API providers for this model.