LLMs ·

LongCat 2.5 (1.6T-A48B) New

Parameters
1,780B (48.0B active)
Architecture
Mixture of experts
Max context
512K tokens
License
mit
Released
2026-11

How to run it

Original weights · Hugging Face

Full-precision files from Meituan, for fine-tuning, vLLM, Transformers.

huggingface.co/meituan-longcat/LongCat-2.5 (opens in a new tab)

No sign-in needed.

Links checked 2 hours ago against Hugging Face, Ollama and GitHub.

Apps that run LongCat 2.5 (1.6T-A48B)

Our pick for each kind of computer, assuming the model fits in its memory. Pick your own GPU or device in the Model Finder for an exact answer.

Mac (Apple silicon)

Docker Model Runner

Handy if you already work in Docker.

Also good:

Windows PC, NVIDIA GPU

Docker Model Runner

Handy if you already work in Docker.

Also good:

Windows PC, AMD Radeon

Docker Model Runner

Handy if you already work in Docker.

Also good:

Linux, NVIDIA GPU

Docker Model Runner

Handy if you already work in Docker.

Also good:

Linux server, 80 GB+ GPUs

Docker Model Runner

Handy if you already work in Docker.

Also good:

App Type Opens it as Systems Engine Start
vLLM Server Original weights WSLWindowsLinux vLLM 
vllm serve meituan-longcat/LongCat-2.5
SGLang Server Original weights WSLWindowsLinux SGLang 
python3 -m sglang.launch_server --model-path meituan-longcat/LongCat-2.5
Docker Model Runner CLI + API Original weights macOSWindowsLinux llama.cpp, vLLM 
docker model run hf.co/meituan-longcat/LongCat-2.5
Text Generation WebUI Desktop app Original weights macOSWindowsLinux llama.cpp, Transformers, ExLlamaV3 Get Text Generation WebUI

Which apps can open a model depends on the files published for it: GGUF for llama.cpp-based apps, MLX for Apple’s engine, the original weights for vLLM and SGLang. Compare every app.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps. Try another context length.

KV cache: latent (MLA) KV cache; about 48B parameters are active per token; the card counts 1.6T, the files hold 1.78T with the n-gram embedding and vision parts.

Quantization Quality Weights Total at 8K context Total at 32K context Download
Q2_K Low 712 GB 713 GB 714 GB All files (opens in a new tab)
Q3_K_M Low 872 GB 873 GB 874 GB All files (opens in a new tab)
Q4_K_M Good 1,086 GB 1,087 GB 1,088 GB All files (opens in a new tab)
Q5_K_M High 1,264 GB 1,265 GB 1,266 GB All files (opens in a new tab)
Q6_K High 1,459 GB 1,460 GB 1,461 GB All files (opens in a new tab)
Q8_0 Very high 1,887 GB 1,887 GB 1,889 GB All files (opens in a new tab)
F16 Full 3,560 GB 3,560 GB 3,562 GB Original weights (opens in a new tab)

Best hardware for LongCat 2.5 (1.6T-A48B)

None. No single GPU or device in our catalog runs this model on its own; it needs several GPUs working together.

Check it on your own hardware

Run it on a cluster

Devices that can’t run LongCat 2.5 (1.6T-A48B) alone can run it together. These example clusters do:

Try it with your own devices

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.

Was this page helpful?