// Meituan LongCat · Text LLM
LongCat-Flash-Thinking: VRAM & local deployment
560B total · 27B active (MoE) parameters. About 374 GB at 4-bit and 1.23 TB at BF16 with one 8K-token request.
released 2025-09-21licence: Commercial OKmemory figures are estimates
01Specs
| Publisher | Meituan · meituan-longcat |
|---|---|
| Family / series | Meituan LongCat · LongCat-Flash-Thinking |
| Type | Text LLM (text-generation) |
| Parameters | 560B total · 27B active (MoE) Source: official model card“560 billion total parameters … # Activated Params 27B” |
| Context length | 128K tokens (131,072)max_position_embeddings in config.json |
| Architecture | 28 layers · MLA · MoE |
| Released weights | BF16 · 1.12 TBdownload |
| Licence | MIT · Commercial OKPermissive licence (Apache-2.0 / MIT / CC-BY / OpenMDW): commercial use allowed. |
| Released | 2025-09-21 (Hugging Face repo created) |
02How much VRAM it needs (estimate)
Total = weights + 10% + 0.5GB overhead + KV cache for one request (FP16 KV). More concurrent users multiply the KV column.
| Precision | Weights | Total @ 8K | Total @ 32K | Total @ 128K | Smallest GPU | Mac |
|---|---|---|---|---|---|---|
| BF16 / FP16 (original) | 1.12 TB | 1.23 TB | 1.23 TB | 1.24 TB | 8 × B200 (1,440GB) | — |
| FP8 / INT8 / Q8_0 | 595 GB | 655 GB | 656 GB | 659 GB | 8 × H200 (1,128GB) | — |
| 4-bit (Q4_K_M / AWQ / GPTQ-Int4) | 340 GB | 374 GB | 375 GB | 378 GB | 8 × H100 (640GB) | — |
KV cache per request: 8K 0.3 GB · 32K 1.1 GB · 128K 4.2 GB
Bits per weight: BF16 16, FP8 8.5, 4-bit 4.85. “Fits” = total ≤ 90% of memory; Mac GPU can use ~2/3 of unified memory (≤36GB) or ~3/4 (larger). For MoE, all experts must be in memory, so size by total parameters. Method →
03How to run it
Only runtimes the publisher or the runtime’s official library names for this model.
- vLLM — mentioned in the official model card
- SGLang — mentioned in the official model card
- Transformers — mentioned in the official model card
04Official repositories & quantised builds
2 official repo(s) from meituan-longcat.
05Other Meituan LongCat models
All 16 →| Model | Params | 4-bit total | Released |
|---|---|---|---|
| LongCat-Flash-Thinking-2601 · same series | 560B / 27B | 374 GB | 2026-01-14 |
| LongCat-Flash-Lite-Sparse | 69B / 3B | 46.6 GB | 2026-07-31 |
| LongCat-2.0 | 1.6T / 48B | 1.07 TB | 2026-07-05 |
| LongCat-AudioDiT-1B | 1.42B | 1.4 GB | 2026-03-30 |
| LongCat-AudioDiT-3.5B | 3.83B | 3.1 GB | 2026-03-30 |
| LongCat-Image-Edit-Turbo | — | — | 2026-02-03 |
| LongCat-Flash-Lite | 68.5B / 3B | 46.3 GB | 2026-01-27 |
| LongCat-Flash-Thinking-ZigZag | 561.9B | 376 GB | 2026-01-23 |
| LongCat-HeavyMode-Summary | 561.9B | 376 GB | 2026-01-14 |
06Sources
Data checked 2025-09-24. Parameters, context and licence are copied from the official repository; memory figures are our estimates.