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...
Anyone in the Project can @-mention Meituan: LongCat 2.0 with the team's shared context - pooled credits, one chat, one memory.
Starter is free forever - 1 Project, 100 credits/month, 1 MCP. No card.
Verdict
Best for
- Processing large Chinese-language documents
- Cost-sensitive multi-document analysis
- Bilingual Chinese-English translation tasks
- High-volume context ingestion on budget
Strengths
The 1.05M token context window rivals Gemini 1.5 Pro while undercutting it by 85% on input costs. Meituan's background in Chinese e-commerce and local services suggests strong training on Chinese text, making this a rare option for teams needing affordable long-context processing in Mandarin. The $1.20 output pricing remains competitive for summarization and extraction tasks where output tokens stay low relative to input volume.
Trade-offs
No public benchmarks means no visibility into reasoning depth, factual accuracy, or instruction-following against established models. Western users will likely see weaker performance on English-only tasks compared to Anthropic or OpenAI offerings. The proprietary license and Meituan's China focus raise questions about data handling for non-Chinese enterprises. Without MMLU, HumanEval, or similar scores, you're testing blind.
Specifications
- Provider
- meituan
- Category
- llm
- Context length
- 1,048,756 tokens
- Max output
- 262,144 tokens
- Modalities
- text
- License
- proprietary
- Released
- 2026-07-20
Pricing
- Input
- $0.30/Mtok
- Output
- $1.20/Mtok
- Model ID
meituan/longcat-2.0
Per-token prices show what the model costs upstream. On Switchy your team draws from one shared org credit pool - one plan, one balance for everyone.
Team cost calculator
5 seats · 80 msgs/day
Switchy meters this against your org's shared credit pool - one plan, one balance for everyone.
Providers
Performance
Benchmarks
Works well with
Top MCPs
Compatibility data comes from first-party telemetry; once we have enough co-usage signal, top MCPs for this model will appear here.
How Switchy teams use it
Starter prompts
Summarize Chinese Legal Documents
Read this Chinese contract in full. Extract all payment terms, liability clauses, and termination conditions. Present findings in a bulleted list with English translations for each key term.Open in a Project →
Cross-Reference Product Reviews
I'm pasting 500 customer reviews for this product from Chinese platforms. Identify the top 5 recurring complaints and the top 3 praised features. Provide counts and representative quotes for each.Open in a Project →
Translate Technical Documentation
Translate this 200-page Chinese technical manual to English. Maintain consistent terminology for all technical terms. Flag any sections where context is ambiguous and translation confidence is low.Open in a Project →
Aggregate Multi-Source Reports
I'm providing 50 Chinese news articles about this policy change. Synthesize them into a 500-word English summary covering: timeline, key stakeholders, economic impact, and public reaction.Open in a Project →
Extract Data From Scanned Forms
Extract all customer names, phone numbers, addresses, and order IDs from these 100 Chinese order forms. Output as a CSV with columns: name, phone, address, order_id. Flag any entries with missing data.Open in a Project →
Compare with
More language models
- Meta: Llama 3.1 70B Instructmeta-llama
- Meta: Llama 3.1 8B Instructmeta-llama
- Meta: Llama 3.2 1B Instructmeta-llama
- Meta: Llama 3.2 3B Instructmeta-llama
- Meta: Llama 3.3 70B Instructmeta-llama
- Meta: Llama 4 Maverickmeta-llama
- Meta: Llama 4 Scoutmeta-llama
- Meta: Llama Guard 4 12Bmeta-llama
- Meta: Muse Glimmer 30Bmeta
- Meta: Muse Glimmer 30B (batch)meta
- Meta: Muse Spark 1.1meta
- Meta: Muse Spark 1.2meta