How much memory does Llama 3.2 1B Instruct need at Q4?

The Q4 figure used here is 0.8 GB. GGUF file size from the linked Hub repo. KV cache was not added, because this model's attention shape is not in the catalog.

What is the cheapest catalog device that runs Llama 3.2 1B Instruct at Q4?

GeForce RTX 5050, list price $249.

How fast is Llama 3.2 1B Instruct on the Framework Desktop 128GB, as a bandwidth estimate?

About 104 tok/s at FP16. That is device bandwidth divided by active weight bytes, not a measured run.

Yes. Llama 3.2 1B Instruct fits on the Framework Desktop 128GB at FP16, with 93.0 GB of headroom. Usable memory is 96.0 GB.

Can the Framework Desktop 128GB run Llama 3.2 1B Instruct?

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

Usable memory on this 128 GB unified-memory config is 96.0 GB (75% of the sticker figure). A result under 2 GB of headroom is labeled tight.

Quant ladder

Estimated memory for Llama 3.2 1B Instruct on Framework Desktop 128GB
Quant	Estimate	Headroom	Est. tok/s	Verdict
Q4	0.8 GB	95.2 GB	317 tok/s	Yes
Q8	1.3 GB	94.7 GB	194 tok/s	Yes
FP16	3.0 GB	93.0 GB	104 tok/s	Yes

GGUF file size from the linked Hub repo. KV cache was not added, because this model's attention shape is not in the catalog.
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

No KV-cache estimate. The public config for this model was gated, missing, or not a GQA or MLA shape this catalog knows how to read.

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 unsloth/Llama-3.2-1B-Instruct-GGUF, the Hub repo whose file size is used above. It is not an official Ollama library tag.

ollama run hf.co/unsloth/Llama-3.2-1B-Instruct-GGUF:Q4_K_M

Other picks

Cheapest catalog device at Q4: GeForce RTX 5050
Closest catalog model that still fits at Q4: Mixtral 8x22B.

All hardware for Llama 3.2 1B Instruct · All models for the Framework Desktop 128GB

Questions

How much memory does Llama 3.2 1B Instruct need at Q4?: The Q4 figure used here is 0.8 GB. GGUF file size from the linked Hub repo. KV cache was not added, because this model's attention shape is not in the catalog.
Can the Framework Desktop 128GB run Llama 3.2 1B Instruct?: Yes. Llama 3.2 1B Instruct fits on the Framework Desktop 128GB at FP16, with 93.0 GB of headroom. Usable memory is 96.0 GB.
What is the cheapest catalog device that runs Llama 3.2 1B Instruct at Q4?: GeForce RTX 5050, list price $249.
How fast is Llama 3.2 1B Instruct on the Framework Desktop 128GB, as a bandwidth estimate?: About 104 tok/s at FP16. That is device bandwidth divided by active weight bytes, not a measured run.