Can the Apple A18 Pro (8 GB) run MOSS-Image SFT?

Barely: it runs very slowly (CPU offload)

Very slow (CPU offload). At Q8 GGUF it needs about 4.1 GB of 5.4 GB usable and takes about 10.1 min per image.

Best precision: Q8 GGUF
Memory mode: Sequential CPU offload
Time per image: 10.1 min
Output: 1024×1024, 50 steps

Best app to run it on the Apple A18 Pro (8 GB)

On macOS

ComfyUI

Desktop app · runs on PyTorch (ComfyUI), GGUF via ComfyUI-GGUF

Runs this model with built-in or popular custom nodes; GGUF files load with ComfyUI-GGUF.

Get ComfyUI

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download MOSS-Image SFT

Recommended for the Apple A18 Pro (8 GB): q8_gguf. Every other precision has its own download link in the table below.

Quantized files · GGUF for ComfyUI

Smaller GGUF versions for ComfyUI (with the ComfyUI-GGUF node). No sign-in needed; the model’s license still applies.

Download the q8_gguf (Q8_0) file (opens in a new tab)

huggingface.co/unsloth/MOSS-Image-SFT-GGUF (opens in a new tab)

Original weights · Hugging Face

Full-precision files from OpenMOSS, for fine-tuning, vLLM, Transformers, Diffusers or ComfyUI.

huggingface.co/OpenMOSS-Team/MOSS-Image-SFT (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from OpenMOSS. The weights themselves download from Hugging Face.

github.com/OpenMOSS/MOSS-Image (opens in a new tab)

git clone https://github.com/OpenMOSS/MOSS-Image

Links checked 2 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 81Great quality (A) · 81 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q8 GGUF, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Very slowCPU offload: How well it runs on the Apple A18 Pro (8 GB): speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.	36.5 / 45
Memory fitFull marks when the model fits in the share of unified memory the GPU can use; fewer when it goes past that limit.	12.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	1.9 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	15.0 / 15
Match score	65.5 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every precision on the Apple A18 Pro (8 GB)

Quant	Memory needed	Available	Speed	Memory mode	Runs	Quality	Download
`FP16`	5.8 GB	5.4 GB	—	—	Won't run	82Great quality (A) · 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 11/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)
`FP8`	4.0 GB	5.4 GB	7.9 min	Sequential CPU offload	Very slowCPU offload	80Great quality (A) · 80 of 100: how capable the model is at FP8, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)
`Q8 GGUF`	4.1 GB	5.4 GB	10.1 min	Sequential CPU offload	Very slowCPU offload	81Great quality (A) · 81 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q4 GGUF`	3.3 GB	5.4 GB	10.5 min	Sequential CPU offload	Very slowCPU offload	76Great quality (A) · 76 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)

Other hardware that runs it

80Great quality (A) · 80 of 100: how capable the model is at FP8, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4090 FP8 · 13 s · Runs great
80Great quality (A) · 80 of 100: how capable the model is at FP8, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. AMD Radeon RX 7900 XTX FP8 · 14 s · Runs well
80Great quality (A) · 80 of 100: how capable the model is at FP8, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5090 FP8 · 11 s · Runs great
80Great quality (A) · 80 of 100: how capable the model is at FP8, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 3090 FP8 · 24 s · Runs well
82Great quality (A) · 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell FP16 · 10 s · Runs great
82Great quality (A) · 82 of 100: how capable the model is at FP16, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. AMD Radeon RX 9070 XT FP16 · 20 s · Runs well (encoder in RAM)

Quick answers

How much memory does MOSS-Image SFT need on the A18 Pro 8GB?

At Q8 GGUF it needs about 4.1 GB. The A18 Pro 8GB has 5.4 GB usable.

How long does MOSS-Image SFT take on the A18 Pro 8GB?

About 10.1 min per image at Q8 GGUF and default settings, an estimate scaled from a reference timing by this hardware’s compute speed.

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.