// Open-model VRAM table · family
Stable Diffusion / Stable Audio: local deployment & VRAM
10 open-weight models from Stability AI (18 official repos incl. quantised builds).
official orgs: stabilityailatest 2026-05-17memory = estimate
01Text LLM (1)
| Model | Params | Context | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|---|
| sv4d2.0 | — | — | — | — | — | — | Commercial with conditions | 2025-02-20 |
02Text-to-speech / audio (5)
| Model | Params | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|
| stable-audio-3-medium | 2.31B | 5.6 GB | 2.0 GB | 8GB GPU | Mac 16GB | Commercial with conditions | 2026-05-17 |
| stable-audio-3-small-music | 568M | 1.7 GB | 0.9 GB | 8GB GPU | Mac 16GB | Commercial with conditions | 2026-05-17 |
| stable-audio-3-small-sfx | 568M | 1.7 GB | 0.9 GB | 8GB GPU | Mac 16GB | Commercial with conditions | 2026-05-17 |
| stable-audio-open-small | 497M | 1.6 GB | 0.8 GB | 8GB GPU | Mac 16GB | Commercial with conditions | 2025-05-12 |
| stable-audio-open-1.0 | 1.21B | 3.2 GB | 1.3 GB | 8GB GPU | Mac 16GB | Commercial with conditions | 2024-05-24 |
03Image generation (4)
| Model | Params | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|
| stable-diffusion-3.5-medium | 2.47B | 5.9 GB | 2.1 GB | 8GB GPU | Mac 16GB | Commercial with conditions | 2024-10-29 |
| stable-diffusion-3.5-large | 8.15B | 18.4 GB | 5.9 GB | 8GB GPU | Mac 16GB | Commercial with conditions | 2024-10-22 |
| stable-diffusion-3.5-large-turbo | 8.15B | 18.4 GB | 5.9 GB | 8GB GPU | Mac 16GB | Commercial with conditions | 2024-10-22 |
| stable-diffusion-3-medium-diffusers | 2.08B | 5.1 GB | 1.9 GB | 8GB GPU | Mac 16GB | Non-commercial | 2024-06-12 |
04Other 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 · 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 · 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 →