modelgrep
M

Meituan: LongCat 2.0

meituan/longcat-2.0

57th smartest of 153ReasoningTools
Use via OpenRouter ↗
Intelligence
33.5
57th of 153
Design Elo
Speed
tokens/sec
Latency
first token
Input price
$0.300
122nd cheapest
Context
1.0M
262K max out

How it compares

Smarter than63%
of all ranked models
Cheaper than60%
of all ranked models

Overview

LongCat 2.0 is a sparse mixture-of-experts language model from Meituan, with 48B active parameters out of 1.6T total. It is suited for coding, repository-level changes, long-horizon problem solving, and agentic...

Benchmarks

independent · Artificial Analysis & Design Arena
Artificial Analysis81th percentile
Intelligence Index
33.5
Coding Index
45.3
GPQA Diamond
78%
Humanity's Last Exam
32%
SciCode
35%

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
AtlasCloudfp8$0.300$1.20100%

Specifications

Context window1.0M
Max output262K
Knowledge cutoff
Input modalitiestext
Output modalitiestext
Prompt caching
Cache read price$0.0060/M
ModeratedNo

LongCat 2.0 FAQ

How much does LongCat 2.0 cost?

LongCat 2.0 costs $0.300 per million input tokens and $1.20 per million output tokens via OpenRouter, making it 122nd cheapest of 307 paid models.

How smart is LongCat 2.0?

LongCat 2.0 scores 33.5 on the Artificial Analysis Intelligence Index, ranking 57th of 153 benchmarked models, with a GPQA Diamond score of 78%.

What is LongCat 2.0's context window?

LongCat 2.0 supports a 1.0M-token context window and can output up to 262K tokens. It accepts text input.

Compare head-to-head