Best image models for an 8 GB Mac
Which open-weight image models run on an 8 GB Mac. Verdicts come from exact Hugging Face file sizes, model card and app maker statements, and published runs.
- Run
- 1
- Unknown
- 1
- Does not fit
- 7
1 of 9 listed image models run on an 8 GB Mac: FLUX.2 [klein]. None of them has an independent quality score yet. No figure covers SDXL Base 1.0 at this size yet. FLUX 3 Image Klein, FLUX.2 [dev], Krea 2 Turbo, MOSS-Image, Qwen-Image-2.1, Stable Diffusion 3.5 Large and Z-Image-Turbo need more memory.
License check before commercial use: FLUX.2 [klein].
Get the weekly local AI newsletter
The Thursday ModelFit newsletter: new local models and what fits your machine.
For: MacBook Air M5 16 GB
You'll also get the weekly ModelFit email. By signing up you agree to our Privacy Policy. Unsubscribe from either email anytime, in one click.
Which image models run on an 8 GB Mac?
9 modelsRuns means a published run on 8 GB or less, or a model card or app maker figure at or below it. Does not fit means the weights the app keeps resident in the smallest complete build are larger than 8 GB. A Mac has one memory pool, so the text encoder cannot move out to a separate system RAM; an app that loads the text encoder and the main model one after the other counts only its largest stage. unknown means no source covers this size yet; it is not a failure.
| Model | Verdict | Why | Quality | App |
|---|---|---|---|---|
| FLUX.2 [klein]Apache 2.0 (4B), FLUX Non-Commercial License (9B) | Runs · Stated | Stated by the source: 8 GB of unified memory runs it (iris.c with memory-mapped weights, Klein 4B).iris.c README | No independent score yet | flux-2-swift-mlx, Diffusers, stable-diffusion.cpp |
| SDXL Base 1.0CreativeML Open RAIL++-M License | unknown | The smallest complete build, Rapid-MLX 8-bit UNet, is a 6.9 GB download, but no stated figure or published run covers 8 GB of unified memory yet. | No independent score yet | ComfyUI, Rapid-MLX, stable-diffusion.cpp |
| FLUX 3 Image KleinFLUX Kommunity License v1.0 | Does not fit est. | The smallest complete build, Q4_K_M GGUF, keeps 12.9 GB of weights resident, more than 8 GB of unified memory. | No independent score yet | Diffusers, mflux |
| FLUX.2 [dev]FLUX Non-Commercial License | Does not fit est. | The smallest complete build, Q4_K_M GGUF, keeps 34.8 GB of weights resident, more than 8 GB of unified memory. | #38 of 82 in text to image77,988 votes | flux-2-swift-mlx, Diffusers, ComfyUI, stable-diffusion.cpp |
| Krea 2 TurboKrea 2 Community License Agreement v.1 | Does not fit est. | The smallest complete build, Turbo Q4_K_M GGUF, keeps 16.9 GB of weights resident, more than 8 GB of unified memory. | No independent score yet | ComfyUI-Krea2-MLX, Diffusers, stable-diffusion.cpp |
| MOSS-ImageApache 2.0 | Does not fit est. | The smallest complete build, UD-Q4_K_XL GGUF, keeps 13.2 GB of weights resident, more than 8 GB of unified memory. | No independent score yet | MOSS-Image MPS path, stable-diffusion.cpp |
| Qwen-Image-2.1Qwen Research License Agreement | Does not fit est. | The smallest complete build, Q4_K_M GGUF, keeps 10.7 GB of weights resident, more than 8 GB of unified memory. | #19 of 82 in text to image11,441 votes | Diffusers with staged loading, Diffusers, stable-diffusion.cpp |
| Stable Diffusion 3.5 LargeStability AI Community License | Does not fit est. | The smallest complete build, Q4_0 GGUF, keeps 16.4 GB of weights resident, more than 8 GB of unified memory. | #81 of 82 in text to image23,396 votes | ComfyUI, Rapid-MLX |
| Z-Image-TurboApache 2.0 | Does not fit est. | The smallest complete build, Q4_K_M GGUF, keeps 12.1 GB of weights resident, more than 8 GB of unified memory. | #59 of 82 in text to image24,808 votes | Diffusers, mflux, stable-diffusion.cpp |
Quality: blind human votes from Arena (arena.ai), CC BY 4.0, snapshot 2026-10-07. Scores are for the Arena checkpoint, not your local quantized build.
Frequently asked questions
What is the best image model for an 8 GB Mac?
FLUX.2 [klein] runs on an 8 GB Mac. None of them has an independent quality score yet.
Which image models run on an 8 GB Mac?
FLUX.2 [klein] (8 GB stated by iris.c README).
Which image models need more than 8 GB on a Mac?
FLUX 3 Image Klein (smallest complete build Q4_K_M GGUF, 12.9 GB), FLUX.2 [dev] (smallest complete build Q4_K_M GGUF, 34.8 GB), Krea 2 Turbo (smallest complete build Turbo Q4_K_M GGUF, 16.9 GB), MOSS-Image (smallest complete build UD-Q4_K_XL GGUF, 13.2 GB), Qwen-Image-2.1 (smallest complete build Q4_K_M GGUF, 10.7 GB), Stable Diffusion 3.5 Large (smallest complete build Q4_0 GGUF, 16.4 GB) and Z-Image-Turbo (smallest complete build Q4_K_M GGUF, 12.1 GB). Their resident weights alone are larger than the memory.