Can DiffusionGemma 26B A4B run on Mac Studio M3 Ultra 256GB?

YES — Runs Great

A75Great

Estimated from fit model

DiffusionGemma 26B A4B needs ~49.4 GB VRAM. Mac Studio M3 Ultra 256GB has 184.3 GB. With Q4_K_M quantization, expect ~67 tok/s.

Runtime: vLLMCapacity: RoomyBandwidth: HighStack: OptimizedBottleneck: Balanced

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 49.4 GB, 66.6 tok/s, Runs well

49.4 GB required184.3 GB available

27% VRAM used

Fit status

Runs well

Decode

66.6 tok/s

TTFT

2907 ms

Safe context

262K

Memory

49.4 GB / 184.3 GB

Memory breakdown

Weights15.7 GB

KV Cache3.7 GB

Runtime2.4 GB

Headroom27.6 GB

See how fast it feels

See how fast it feelsDiffusionGemma 26B A4B on Mac Studio M3 Ultra 256GB
1st promptCold start — includes initialization
>Solve: If a train travels at 60 mph for 2.5 hours, how far does it go?
Let me work through this step by step.

Given information:
  - Speed = 60 mph (miles per hour)
  - Time  = 2.5 hours

Formula: Distance = Speed x Time

Calculation:
  Distance = 60 mph x 2.5 h
           = 150 miles

The train travels 150 miles.

Double-check: at 60 mph the train covers 60 miles each hour. In half an hour it covers 30 miles. So in 2.5 hours: 60 + 60 + 30 = 150 miles. Confirmed.
2nd promptWarm — model ready, no init overhead
>What if the train stops for 20 minutes halfway?
The stop doesn't change the total distance covered while moving, but it affects total trip time.

Breakdown:
  - First half: 1.25 hours at 60 mph = 75 miles
  - Stop: 20 minutes (no distance)
  - Second half: 1.25 hours at 60 mph = 75 miles

Total distance: 150 miles (same)
Total time: 2.5 h + 0.33 h = 2.83 hours
Average speed: 150 / 2.83 = 53 mph

The distance stays the same but average speed drops to 53 mph because of the stop.
Estimated: 66.6 tok/s decode · 2.9s TTFT (warm) · 167 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

Shared-memory contention still exists

The OS, browser, and inference runtime all compete for the same physical memory pool, so real-world headroom is less forgiving than raw capacity suggests.

Best improvement path

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	A	Runs well	66.6 tok/s	1586 ms	262K
Coding	A	Runs well	66.6 tok/s	2907 ms	262K
Agentic Coding	A	Runs well	66.6 tok/s	4228 ms	262K
Reasoning	A	Runs well	66.6 tok/s	3436 ms	262K
RAG	A	Runs well	66.6 tok/s	5286 ms	262K

Inference speed

DiffusionGemma 26B A4B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for DiffusionGemma 26B A4B at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~144 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.

GPU / Mac	Memory	Quant	Speed (tok/s)	Fits?
RTX 5090 32GB	32 GB	Q4_K_M	143.6	Fits
RTX 4090 24GB	24 GB	Q4_K_M	67.5	Too big
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	66.6	Fits
RX 7900 XTX 24GB	24 GB	Q4_K_M	60.9	Too big
RTX 3090 24GB	24 GB	Q4_K_M	57.7	Too big
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	55.5	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	52.6	Fits
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	41.1	Fits
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	41.1	Fits
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	28.7	Fits
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	26.3	Fits
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	25.1	Fits
RTX 4080 Super 16GB	16 GB	Q4_K_M	24.6	Too big
RTX 4070 12GB	12 GB	Q4_K_M	8.6	Too big
RTX 3060 12GB	12 GB	Q4_K_M	5.4	Too big
RTX 4060 8GB	8 GB	Q4_K_M	3.6	Too big

Estimates for single-stream decoding at Q4_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.

Quantization options

How DiffusionGemma 26B A4B (25.799999237060547B params) fits at each quantization level on Mac Studio M3 Ultra 256GB (184.3 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	10.1 GB	Low	B66
Q3_K_S	3	12.6 GB	Low	B66
NVFP4	4	14.4 GB	Medium	B66
Q4_K_M	4	15.7 GB	Medium	B66
Q5_K_M	5	18.6 GB	High	B66
Q6_K	6	21.2 GB	High	B66
Q8_0	8	27.6 GB	Very High	B67
F16Best for your GPU	16	52.9 GB	Maximum	A70

Get started

Copy-paste commands to run DiffusionGemma 26B A4B on your machine.

Run

docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
  --hf-repo "google/diffusiongemma-26B-A4B-it" \
  --hf-file "diffusiongemma-26B-A4B-it-Q4_K_M.gguf" \
  -c 4096 -ngl 99

Your hardware

More models your Mac Studio M3 Ultra 256GB can run

Model	Params	Grade	Decode
Devstral 2 123B Instruct	123B	S	6.5 tok/s
Qwen3-Coder 30B A3B Instruct	30.5B	S	64 tok/s
Qwen 3.5 27B	27B	S	29.2 tok/s
Qwen 3.6 27B	27B	S	29.3 tok/s
Qwen 3.5 122B A10B	122B	S	17 tok/s

Frequently asked questions

Can Mac Studio M3 Ultra 256GB run DiffusionGemma 26B A4B?

Yes, Mac Studio M3 Ultra 256GB can run DiffusionGemma 26B A4B with a A grade (Runs well). Expected decode speed: 66.6 tok/s.

How much VRAM does DiffusionGemma 26B A4B need?

DiffusionGemma 26B A4B (25.799999237060547B parameters) requires approximately 49.4 GB of memory with Q4_K_M quantization.

What is the best quantization for DiffusionGemma 26B A4B?

The recommended quantization for DiffusionGemma 26B A4B is Q4_K_M, which balances quality and memory efficiency.

What speed will DiffusionGemma 26B A4B run at on Mac Studio M3 Ultra 256GB?

On Mac Studio M3 Ultra 256GB, DiffusionGemma 26B A4B achieves approximately 66.6 tokens per second decode speed with a time-to-first-token of 2907ms using Q4_K_M quantization.

Can Mac Studio M3 Ultra 256GB run DiffusionGemma 26B A4B for coding?

For coding workloads, DiffusionGemma 26B A4B on Mac Studio M3 Ultra 256GB receives a A grade with 66.6 tok/s and 262K context.

What context window can DiffusionGemma 26B A4B use on Mac Studio M3 Ultra 256GB?

On Mac Studio M3 Ultra 256GB, DiffusionGemma 26B A4B can safely use up to 262K tokens of context. The model's official context limit is 262K, but available memory constrains the safe maximum.

Is unified memory on Mac Studio M3 Ultra 256GB as fast as VRAM for DiffusionGemma 26B A4B?

Not always. Mac Studio M3 Ultra 256GB can often fit larger models thanks to unified memory, but a discrete GPU with dedicated high-bandwidth VRAM may still decode faster once the model fits. For this combination, the important distinction is capacity versus sustained throughput.

See all results for Mac Studio M3 Ultra 256GB See all hardware for DiffusionGemma 26B A4B

Embed this result▼

Paste this snippet into any page to show a live fit card.

<iframe src="https://willitrunai.com/embed/diffusiongemma-26b-a4b-on-m3-ultra-256gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: