LLM Reference

LongCat-2.0 on LongCat API Platform

LongCat · Meituan

ServerlessOpen Source

Last refreshed 2026-07-07. Next refresh: weekly.

Why use LongCat-2.0 on LongCat API Platform?

LongCat API Platform offers LongCat-2.0 with pay-as-you-go pricing at $0.75/1M input tokens. LongCat API Platform is Meituan's public model API service.

Input / 1M
$0.75
Output / 1M
$2.95
Cache
read $0.015
Batch
Not sourced

Setup recipe

Docs fallback
Install
Use the provider REST API or SDK
Auth
Create a provider API key
Call
model: LongCat-2.0
Model ID
LongCat-2.0

Request example

Curated snippets for this provider are not sourced yet. Use LongCat API Platform documentation with model ID LongCat-2.0.

Gotchas

  • Use provider model ID "LongCat-2.0", not the LLMReference slug "longcat-2-0".

Pricing

TypePrice (per 1M)
Input tokens$0.75
Output tokens$2.95

Capabilities

ReasoningFunction CallingTool UsePrompt Caching

About LongCat-2.0

LongCat-2.0 is Meituan's large-scale MoE language model for coding, long-context, and agentic tasks. The model has 1.6T total parameters, roughly 48B activated per token, native 1M context support, and API access through OpenAI- and Anthropic-compatible LongCat endpoints. Public MIT-licensed safetensors weights are downloadable from Hugging Face, with Transformers, vLLM, and SGLang deployment support.

FAQ

What does LongCat-2.0 cost on LongCat API Platform?

On LongCat API Platform, LongCat-2.0 costs $0.75 per 1M input tokens and $2.95 per 1M output tokens.

What is the context window for LongCat-2.0 on LongCat API Platform?

LongCat-2.0 supports a 1m token context window on LongCat API Platform.

What API model ID do I use for LongCat-2.0 on LongCat API Platform?

Use the model ID LongCat-2.0 when calling LongCat API Platform's API.

Who created LongCat-2.0?

LongCat-2.0 was created by Meituan as part of the LongCat model family.

Is LongCat-2.0 open source?

LongCat-2.0 is open source under MIT according to the seed data.

Get Started

Model Specs

Released2026-06-30
Parameters1.6T total; ~48B active
Context1m
ArchitectureMixture of Experts