How much memory does Qwen3 8B need at Q4?

The Q4 figure used here is 6.2 GB. GGUF file size plus an FP16 KV-cache estimate at 8,192 tokens. The cache counts only the layers and heads named in the method on the About page.

What is the cheapest catalog device that runs Qwen3 8B at Q4?

GeForce RTX 5050, list price $249.

How fast is Qwen3 8B on the Mac Studio M3 Ultra 96GB, as a bandwidth estimate?

About 50.0 tok/s at FP16. That is device bandwidth divided by active weight bytes, not a measured run.

Yes. Qwen3 8B fits on the Mac Studio M3 Ultra 96GB at FP16, with 47.5 GB of headroom. Usable memory is 67.2 GB.

Can the Mac Studio M3 Ultra 96GB run Qwen3 8B?

Data updated 16 October 2026. Last reviewed the same day, when this catalog snapshot was written.

Usable memory on this 96 GB unified-memory config is 67.2 GB (70% of the sticker figure). A result under 2 GB of headroom is labeled tight.

Quant ladder

Estimated memory for Qwen3 8B on Mac Studio M3 Ultra 96GB
Quant	Estimate	Headroom	Est. tok/s	Verdict
Q4	6.2 GB	61.0 GB	163 tok/s	Yes
Q8	9.9 GB	57.3 GB	94.0 tok/s	Yes
FP16	19.7 GB	47.5 GB	50.0 tok/s	Yes

GGUF file size plus an FP16 KV-cache estimate at 8,192 tokens. The cache counts only the layers and heads named in the method on the About page.
Formula estimate: parameter count × bytes per weight × 1.20. No GGUF file size is stored for this quant.
Tokens/sec is a bandwidth roofline: GB/s divided by active weight GB. It is not a measured generation speed.

KV cache at 8K and 32K

FP16 KV-cache estimate: 1.2 GB at 8,192 tokens and 4.8 GB at 32,768 tokens. GQA counts two tensors (K and V) for the stored KV heads and head dimension.

Offload

This machine has one memory pool. Weights are not offloaded to a second bank of system RAM.

Ollama

This command pulls the Q4_K_M file from Qwen/Qwen3-8B-GGUF, the Hub repo whose file size is used above. It is not an official Ollama library tag.

ollama run hf.co/Qwen/Qwen3-8B-GGUF:Q4_K_M

Other picks

Cheapest catalog device at Q4: GeForce RTX 5050
Closest catalog model that still fits at Q4: Llama 4 Scout.

All hardware for Qwen3 8B · All models for the Mac Studio M3 Ultra 96GB

Questions

How much memory does Qwen3 8B need at Q4?: The Q4 figure used here is 6.2 GB. GGUF file size plus an FP16 KV-cache estimate at 8,192 tokens. The cache counts only the layers and heads named in the method on the About page.
Can the Mac Studio M3 Ultra 96GB run Qwen3 8B?: Yes. Qwen3 8B fits on the Mac Studio M3 Ultra 96GB at FP16, with 47.5 GB of headroom. Usable memory is 67.2 GB.
What is the cheapest catalog device that runs Qwen3 8B at Q4?: GeForce RTX 5050, list price $249.
How fast is Qwen3 8B on the Mac Studio M3 Ultra 96GB, as a bandwidth estimate?: About 50.0 tok/s at FP16. That is device bandwidth divided by active weight bytes, not a measured run.