How fast is DeepSeek V4 Flash on an Mac?

Around 4 tokens per second. Token generation on Apple Silicon is limited by memory bandwidth, and this machine has 800 GB/s.

How much memory does DeepSeek V4 Flash need?

166.0 GiB for the weights at Q4_K_M, plus 0.7 GiB for an 8K-token KV cache and about 9.1 GiB of runtime overhead — 175.7 GiB in total.

Home › DeepSeek V4 Flash › M2 Ultra 192GB

Can an M2 Ultra 192GB Mac run DeepSeek V4 Flash?

M2 Ultra · 192 GB unified memory · 800 GB/s · Mac Studio (M2 Ultra), Mac Pro (M2 Ultra)

Just barely — and you will feel it.

DeepSeek V4 Flash at Q4_K_M needs about 175.7 GiB against roughly 176.0 GiB usable — that is 100% of your budget. It will load, but close your browser first, keep the context short, and expect memory pressure. Around 4 tokens/sec.

Weights

166.0 GiB

290.94B params @ Q4_K_M

KV cache

0.7 GiB

8K context, FP16

Overhead

9.1 GiB

runtime + activations

Total needed

175.7 GiB

of ~176.0 GiB usable

Est. speed

4 tok/s

800 GB/s bandwidth

Max context

8K tokens

at Q4_K_M

Where the memory goes

Three things occupy memory when a model is loaded: the weights themselves, the KV cache that holds the conversation, and the runtime's own working buffers. Only the first is fixed — the KV cache grows with every token in your context.

Quantisation	Weights	Total	Tok/s	Fits?
FP16	541.9 GiB	570.5 GiB	1	No
Q8_0	287.9 GiB	303.8 GiB	2	No
Q6_K	223.5 GiB	236.2 GiB	3	No
Q5_K_M	193.1 GiB	204.2 GiB	3	No
Q4_K_M	166.0 GiB	175.7 GiB	4	Yes, but tight
Q3_K_M	132.1 GiB	140.2 GiB	5	Yes, comfortably
Q2_K	101.6 GiB	108.2 GiB	6	Yes, comfortably

The default. Best quality-per-gigabyte for most people.

How long a context fits

The KV cache is often what breaks a setup that looked fine at load time. Quantising it to 8-bit roughly halves its footprint with little measurable quality cost.

Context	KV cache	Total	FP16 KV	Q8 KV
2K	0.2 GiB	175.0 GiB	Yes, but tight	Yes, but tight
4K	0.3 GiB	175.2 GiB	Yes, but tight	Yes, but tight
8K	0.7 GiB	175.7 GiB	Yes, but tight	Yes, but tight
16K	1.3 GiB	176.7 GiB	No	No
32K	2.7 GiB	178.6 GiB	No	No
128K	10.8 GiB	190.3 GiB	No	No

What to run instead on an M2 Ultra 192GB Mac

These are the largest models that fit comfortably on this machine at Q4_K_M with an 8K context.

Model	Size	Total	Tok/s
gpt-oss-120b	116.8B	71.3 GiB	128	Yes, comfortably
Qwen2.5 72B	72.7B	46.8 GiB	14	Yes, comfortably
DeepSeek-R1-Distill-Llama-70B	70.6B	45.6 GiB	15	Yes, comfortably
Llama 3.3 70B	70.6B	45.6 GiB	15	Yes, comfortably
Mixtral 8x7B	46.7B	29.8 GiB	51	Yes, comfortably
DeepSeek-R1-Distill-Qwen-32B	32.8B	22.4 GiB	32	Yes, comfortably

How to run it

Ollama is the shortest path. It picks a quantisation automatically — usually Q4_K_M, which is what the numbers above assume.

brew install ollama
ollama serve &
ollama run deepseek-v4-flash

For MLX — Apple's own array framework, typically a little faster than llama.cpp on Apple Silicon:

pip install mlx-lm
mlx_lm.generate --model mlx-community/DeepSeek-V4-Flash-4bit --prompt "Hello"

You are close to the limitIf it stutters or gets killed, raise the Metal memory cap before loading: sudo sysctl iogpu.wired_limit_mb=167116. This resets on reboot. Quantising the KV cache (--kv-cache-type q8_0 in llama.cpp) buys back 0.3 GiB.

Macs that run DeepSeek V4 Flash comfortably

The smallest configuration that handles it is the M3 Ultra 256GB (Mac Studio (M3 Ultra)).

Mac	Memory	Bandwidth	Tok/s
M3 Ultra 256GB	256 GB	819 GB/s	4
M3 Ultra 512GB	512 GB	819 GB/s	4

Common questions

Can an M2 Ultra 192GB Mac run DeepSeek V4 Flash?: Yes. At Q4_K_M it needs about 175.7 GiB of the roughly 176.0 GiB available, generating around 4 tokens per second.
How fast is DeepSeek V4 Flash on an Mac?: Around 4 tokens per second. Token generation on Apple Silicon is limited by memory bandwidth, and this machine has 800 GB/s.
How much memory does DeepSeek V4 Flash need?: 166.0 GiB for the weights at Q4_K_M, plus 0.7 GiB for an 8K-token KV cache and about 9.1 GiB of runtime overhead — 175.7 GiB in total.