Meituan · Text
LongCat models: Versions, specs, GPU requirements & hosting
Meituan LongCat 2.0 is a frontier-scale sparse model for long-horizon coding and agentic tasks with official GPU and NPU deployment guidance.
Overview and history
LongCat is an official Meituan model family represented here through independently verified release, checkpoint, license, runtime and hardware evidence. Meituan LongCat 2.0 is a frontier-scale sparse model for long-horizon coding and agentic tasks with official GPU and NPU deployment guidance. Current public releases are grouped by the General, Flash tracks so sibling sizes and specialist branches are not presented as false chronological generations.
FusionPointAI currently tracks 5 certified public LongCat releases. The catalogue preserves release dates and evaluates Current, Recent, Previous and Legacy status within each model track rather than by database insertion order.
Sparse 1.6T scale
Long-context training
SGLang deployment path
Architecture and model range
The verified public range spans 69B to 1600B total parameters and context windows up to 1,048,576 tokens. Each release states whether it is dense or mixture-of-experts and records active parameters separately when the official source publishes them.
Public releases5
Model tracks2
Current generationLongCat Flash Lite Sparse
Last verified2026-09-04T00:00:00.000000Z
Generations, tracks and variants
General
longcat-2-0
Recent generation 3 sibling variantsLongCat-2.0-FP8
Not published by manufacturer parameters · N/A / Not directly comparable context
Quantization
LongCat-2.0
Not published by manufacturer parameters · N/A / Not directly comparable context
Base
LongCat-2.0-INT8
Not published by manufacturer parameters · N/A / Not directly comparable context
Quantization
LongCat 2.0
Current generationFlash
LongCat Flash Lite
Current generationDeployment ecosystem
Choose a specific LongCat release and certified checkpoint before sizing hardware. FusionPointAI rejects unknown runtime configurations, calculates resident memory from total model weights, then distinguishes globally compatible GPUs from currently rentable inventory.
Validate the exact checkpoint license, runtime version, precision kernel, context length, concurrency and multi-GPU topology. Active MoE parameters affect compute, but resident model memory continues to use total stored weights.
Common use cases
Coding agents
Long-horizon automation
Large-cluster inference
License overview
Licenses are certified per release, not inferred from the provider. Review the release page before commercial deployment because terms can differ between generations and variants.
Commercial-use status is a source-backed catalogue classification, not legal advice. Follow the linked official license and attribution requirements.
Release timeline
meituan-longcat/longcat/longcat-20-fp8 @ 4f4d60a60774801748776b8f7fcb6fb0ec177a70907cadb31bbd713bb2466df2
meituan-longcat/longcat/longcat-20 @ 89f5612d7a473d37922c0d4841ef1d9264a49fe398c7b11fd90c9bcc16c6780c
meituan-longcat/longcat/longcat-flash-lite-sparse @ 84a85c4a0529f4faea6ae9cdbef855d40ed557e6afe5f1449b1936c1f4af54d6
meituan-longcat/longcat/longcat-2 @ 413db86e4b84b1f8ea0026d6a0f24fe09928c10334caba6ae58a605bb6b10caf
meituan-longcat/LongCat-2.0 @ 2026-07-08T12:21:34Z
meituan-longcat/LongCat-2.0-INT8 @ 2026-07-08T12:21:25Z
Frequently asked questions
Which LongCat release is current?
Current releases are LongCat 2.0, LongCat Flash Lite Sparse, evaluated within their respective tracks.
Can LongCat be self-hosted?
Only releases with a verified official checkpoint and supported runtime configuration are presented as deployable.
Which GPU should run LongCat?
GPU selection depends on the exact variant, checkpoint precision, context length and concurrency; use the release hardware recommendations rather than the family name alone.
Official references
Last verified: 2026-09-04