M
meituan

Meituan: LongCat 2.0

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...

Overview

Model specifications

Context
1,048,756 tokens
Maximum output
262,144 tokens
Architecture
text->text
Tokenizer
Other
Knowledge cutoff
Not provided
Moderated
No
Capabilities

InputOutput

Input
text
Output
text
Reasoning

Yes

API

Supported API parameters

frequency_penaltyinclude_reasoninglogit_biasmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyseedstoptemperaturetool_choicetoolstop_ktop_p
Available providers

1 Provider

AtlasCloud

fp8
Context
1.0M
Maximum output
262K
Input
$0.300
Output
$1.20
Cache read
$0.0060
Cache write
Not provided