// Open-model VRAM table · family
Ant Ling / Ring / Ming: local deployment & VRAM
29 open-weight models from Ant Group inclusionAI (54 official repos incl. quantised builds). At 4-bit they range from 1.0 GB (Ling-3.0-tiny-UBTree) to 685 GB (Ring-2.6-1T).
official orgs: inclusionAIlatest 2026-10-21memory = estimate
01Text LLM (20)
| Model | Params | Context | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|---|
| Ling-3.0-tiny-UBTree | 183M | 128K | 1.2 GB | 1.0 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-10-21 |
| Ling-3.0-flash-UBTree | 482M | 128K | 2.2 GB | 1.5 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-10-14 |
| Ling-3.1-flash | 560B / 25B | 256K | 1.23 TB | 375 GB | 8 × H100 (640GB) | — | Commercial OK | 2026-10-13 |
| Ling-3.0-tiny | 7.9B / 1.3B | 128K | 18.1 GB | 6.0 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-08-10 |
| Ling-3.0-flash | 124B / 5.1B | 256K | 274 GB | 83.6 GB | RTX PRO 6000 96GB | Mac 128GB | Commercial OK | 2026-08-02 |
| Ring-2.6-1T | 1.03T | 128K | 2.26 TB | 685 GB | 8 × H200 (1,128GB) | — | Commercial OK | 2026-05-14 |
| Ling-2.6-1T | 1.03T | 256K | 2.26 TB | 685 GB | 8 × H200 (1,128GB) | — | Commercial OK | 2026-04-29 |
| Ling-2.6-flash | 104B / 7.4B | 128K | 230 GB | 70.2 GB | H100 80GB | Mac 128GB | Commercial OK | 2026-04-28 |
| Ling-2.5-1T | 1T / 63B | 128K | 2.20 TB | 668 GB | 8 × H200 (1,128GB) | — | Commercial OK | 2026-02-14 |
| Ring-2.5-1T | 1.01T | 128K | 2.23 TB | 676 GB | 8 × H200 (1,128GB) | — | Commercial OK | 2026-02-10 |
| Ring-1T | 1T / 50B | 64K | 2.20 TB | 670 GB | 8 × H200 (1,128GB) | — | Commercial OK | 2025-10-10 |
| Ling-1T | 1T / 50B | 32K | 2.20 TB | 670 GB | 8 × H200 (1,128GB) | — | Commercial OK | 2025-10-02 |
| Ring-flash-linear-2.0 | 104.2B / 6.1B | 128K | 230 GB | 70.5 GB | H100 80GB | Mac 128GB | Commercial OK | 2025-09-25 |
| Ring-mini-linear-2.0 ↗ | 16.4B / 1.6B | 128K | 36.9 GB | 11.8 GB | 16GB GPU | Mac 24GB | Commercial OK | 2025-09-24 |
| Ring-flash-2.0 | 100B / 6.1B | 32K | 221 GB | 67.7 GB | H100 80GB | Mac 128GB | Commercial OK | 2025-09-19 |
| Ling-flash-2.0 | 100B / 6.1B | 32K | 221 GB | 67.7 GB | H100 80GB | Mac 128GB | Commercial OK | 2025-09-17 |
| Ling-mini-2.0 | 16.3B / 1.43B | 32K | 36.6 GB | 11.7 GB | 16GB GPU | Mac 24GB | Commercial OK | 2025-09-08 |
| Ling-lite-1.5 ↗ | 16.8B / 2.75B | 32K | 37.9 GB | 12.2 GB | 16GB GPU | Mac 24GB | Commercial OK | 2025-05-11 |
| Ling-lite ↗ | 16.8B / 2.75B | 32K | 37.9 GB | 12.2 GB | 16GB GPU | Mac 24GB | Commercial OK | 2025-02-28 |
| Ling-plus | 290B / 28.8B | 16K | 642 GB | 197 GB | 4 × RTX PRO 6000 (384GB) | Mac 512GB | Commercial OK | 2025-02-28 |
02Vision-language (1)
| Model | Params | Context | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|---|
| Ling-3.0-flash-VL | 124B / 5.5B | 128K | 274 GB | 83.6 GB | RTX PRO 6000 96GB | Mac 128GB | Commercial OK | 2026-09-04 |
03Omni / audio LLM (2)
| Model | Params | Context | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|---|
| Ming-flash-omni-2.0 | 100B / 6B | 32K | 221 GB | 67.7 GB | H100 80GB | Mac 128GB | Commercial OK | 2026-02-10 |
| Ming-flash-omni-Preview | 104.4B / 6B | 32K | 231 GB | 70.6 GB | H100 80GB | Mac 128GB | Commercial OK | 2025-10-14 |
04Speech-to-text (1)
| Model | Params | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|
| Ming-3.0-asr-tiny | 8.87B | 20.0 GB | 6.4 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-10-14 |
05Text-to-speech / audio (4)
| Model | Params | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|
| Ming-3.0-tts-0.5B | 1.48B | 3.8 GB | 1.5 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-09-22 |
| Ming-3.0-tts-tiny | 8.98B / 1.3B | 20.3 GB | 6.5 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-09-22 |
| Ming-3.0-audio-flash | 126B / 6.5B | 278 GB | 84.5 GB | RTX PRO 6000 96GB | Mac 128GB | Commercial OK | 2026-09-03 |
| Ming-omni-tts-0.5B | 1.62B | 4.1 GB | 1.6 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-02-11 |
06Image generation (1)
| Model | Params | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|
| Ming-Image-0.1-Design | 6.16B | 14.0 GB | 4.6 GB | 8GB GPU | Mac 16GB | Commercial OK | 2026-09-17 |
07Other families
Qwen · IBM Granite · NVIDIA Nemotron · Gemma · Mistral / Magistral / Devstral · GLM / CogVideoX (Z.ai) · DeepSeek · Ai2 OLMo / Molmo · Liquid LFM · MiniCPM / VoxCPM · Tencent Hunyuan / Hy · Phi · Llama · Meituan LongCat · 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 →