// Open-model VRAM table · family
Meituan LongCat: local deployment & VRAM
16 open-weight models from Meituan (24 official repos incl. quantised builds). At 4-bit they range from 46.3 GB (LongCat-Flash-Lite) to 1.07 TB (LongCat-2.0).
official orgs: meituan-longcatlatest 2026-07-31memory = estimate
01Text LLM (8)
| Model | Params | Context | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|---|
| LongCat-Flash-Lite-Sparse ↗ | 69B / 3B | 960K | 152 GB | 46.6 GB | 2 × RTX 5090 (64GB) | Mac 128GB | Commercial OK | 2026-07-31 |
| LongCat-2.0 | 1.6T / 48B | 256K | 3.52 TB | 1.07 TB | 8 × B200 (1,440GB) | — | Commercial OK | 2026-07-05 |
| LongCat-Flash-Lite | 68.5B / 3B | 320K | 151 GB | 46.3 GB | 2 × RTX 5090 (64GB) | Mac 128GB | Commercial OK | 2026-01-27 |
| LongCat-Flash-Thinking-ZigZag | 561.9B | 960K | 1.24 TB | 376 GB | 8 × H100 (640GB) | — | Commercial OK | 2026-01-23 |
| LongCat-Flash-Thinking-2601 | 560B / 27B | 960K | 1.23 TB | 374 GB | 8 × H100 (640GB) | — | Commercial OK | 2026-01-14 |
| LongCat-HeavyMode-Summary | 561.9B | 960K | 1.24 TB | 376 GB | 8 × H100 (640GB) | — | Commercial OK | 2026-01-14 |
| LongCat-Flash-Thinking | 560B / 27B | 128K | 1.23 TB | 374 GB | 8 × H100 (640GB) | — | Commercial OK | 2025-09-21 |
| LongCat-Flash-Chat | 560B / 27B | 128K | 1.23 TB | 374 GB | 8 × H100 (640GB) | — | Commercial OK | 2025-08-29 |
02Omni / audio LLM (1)
| Model | Params | Context | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|---|
| LongCat-Flash-Omni | 560B / 27B | 128K | 1.23 TB | 374 GB | 8 × H100 (640GB) | — | Commercial OK | 2025-10-23 |
03Text-to-speech / audio (2)
| Model | Params | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|
| LongCat-AudioDiT-1B | 1.42B | 3.6 GB | 1.4 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-03-30 |
| LongCat-AudioDiT-3.5B ↗ | 3.83B | 8.9 GB | 3.1 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-03-30 |
04Image generation (4)
| Model | Params | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|
| LongCat-Image-Edit-Turbo | — | — | — | — | — | Commercial OK | 2026-02-03 |
| LongCat-Image-Edit | — | — | — | — | — | Commercial OK | 2025-12-05 |
| LongCat-Image | — | — | — | — | — | Commercial OK | 2025-12-04 |
| LongCat-Image-Dev ↗ | — | — | — | — | — | Commercial OK | 2025-12-04 |
05Video generation (1)
| Model | Params | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|
| LongCat-Video | — | — | — | — | — | Commercial OK | 2025-10-24 |
06Other families
Qwen · IBM Granite · NVIDIA Nemotron · Gemma · Mistral / Magistral / Devstral · GLM / CogVideoX (Z.ai) · DeepSeek · Ai2 OLMo / Molmo · Ant Ling / Ring / Ming · Liquid LFM · MiniCPM / VoxCPM · Tencent Hunyuan / Hy · Phi · Llama · Jina embeddings & rerankers · Wan · Kimi · Cohere Command / Aya · Xiaomi MiMo · FLUX · Microsoft VibeVoice / Fara / Harrier · InternLM / Intern-S · StepFun Step · MiniMax · Baidu ERNIE · NVIDIA Parakeet / Canary speech · Stable Diffusion / Stable Audio · Yi (01.AI) · ByteDance Seed · SmolLM / SmolVLM · BGE embeddings & rerankers · LTX Video · FunAudioLLM speech · gpt-oss · poolside Laguna · Whisper · Baichuan · Thinking Machines Inkling · xAI Grok · rednote dots
Totals include 10% + 0.5GB overhead and one 8K-token KV cache for text models. Estimates only. Method →