# Meituan: LongCat 2.0

Provider: meituan  
Category: llm  
Model ID: `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...

## Specs

- Context length: 1048756 tokens
- Max output: 262144 tokens
- Modalities: text
- Released: 2026-07-20

## Pricing

- Input: $0.30 per million tokens
- Output: $1.20 per million tokens

## Verdict

LongCat 2.0 delivers a 1M+ token context window at aggressive pricing — $0.30 input makes it one of the cheapest options for massive document ingestion. The model targets Chinese-English bilingual tasks, where Meituan's training data gives it an edge over Western-focused alternatives. Trade-off: zero public benchmarks mean you're flying blind on reasoning quality and instruction-following compared to Claude or GPT-4. Best for teams with Chinese language needs and large-context workloads where cost trumps proven performance.

---
Last verified: 2026-09-04T02:00:56.012Z  
Canonical URL: https://switchy.build/directory/models/longcat-2-0