Yes — with 2.1 GB to spare

Llama 3.2 3B Instruct at Q4_K_M fits your M2 · 8 GB entirely in unified memory at 8K context, at an estimated 31 tokens per second. Past 27K the KV cache pushes it over — quantise the cache to q8_0, or step down a quantisation, to go longer.

Fully in unified memory 8K context Q4_K_M · 1.8 GB Llama 3.2 Community Released Sep 2024

The 2024 "it just runs" model for 8 GB laptops. Qwen3.5 4B does the same job better now.

Hardware requirements What hardware do I need to run Llama 3.2 3B Instruct? →

At Q4_K_M with an 8K context: a 6 GB GPU (RTX 3060 Laptop) or a 8 GB Mac (M3). Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.

The VRAM budget

weights 1.8 GB
Weights 1.8 GB KV cache @ 8K 0.88 GB Runtime overhead 0.6 GB Free 2.1 GB of 5.4 GB

Quantisation ladder

QuantWeightsTotal @ 8KMax contextTok/sQualityFit
F16 6.0 GB 7.5 GB — ~9.3 Reference 2.1 GB over
Q8_0 3.2 GB 4.7 GB 14K 18 −0.1% ppl Fits
Q6_K 2.5 GB 3.9 GB 21K 23 −0.4% ppl Long context
Q5_K_M 2.1 GB 3.6 GB 24K 26 −0.8% ppl Long context
Q4_K_M 1.8 GB 3.3 GB 27K 31 −1.9% ppl Recommended
Q3_K_M 1.5 GB 2.9 GB 30K 38 −5.4% ppl Long context
Q2_K 1.3 GB 2.7 GB 32K 45 −15% ppl Long context

Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it.

How to run it

terminal
$ pip install mlx-lm
$ mlx_lm.generate --model mlx-community/Llama-3.2-3B-Instruct-4bit \
    --max-tokens 512 --prompt "Hello"

Apple's own array framework. The fastest path on Apple Silicon. More on MLX.

01Download is 1.8 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02macOS caps what the GPU may wire down at about 5.4 GB of 8 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.
03There is room to go to 27K context on this card.
See all models for this rig Compare with another model

Related models on an M2 · 8 GB

ModelTotalTok/s est.
Llama 3.1 8B Instruct
8.03B · same family
Q4_K_M · 6.1 GB — Won't fit
Llama 3.2 1B Instruct
1.24B · same family
Q4_K_M · 1.5 GB 80 Runs great
Llama 3.3 70B Instruct
70.6B · same family
Q4_K_M · 42.8 GB — Won't fit
Llama 4 Scout 109B-A17B
109B MoE · same family
Q4_K_M · 63.4 GB — Won't fit
Granite 4.1 3B
3.4B · similar size
Q4_K_M · 3.1 GB 29 Runs great
LFM2.5 2.6B
2.7B · similar size
Q4_K_M · 2.2 GB 37 Runs great
Phi-4-mini 3.8B
3.84B · similar size
Q4_K_M · 3.8 GB 26 Runs great

Every model against the M2 · 8 GB →

Llama 3.2 3B Instruct on other hardware

DeviceTotalTok/s est.
GeForce RTX 3060 Laptop
5.3 GB usable · 336 GB/s
3.3 GB 113 Runs great
GeForce RTX 4060 Laptop
7.0 GB usable · 256 GB/s
3.3 GB 86 Runs great
M3 · 8 GB
5.4 GB usable · 100 GB/s
3.3 GB 31 Runs great
GeForce RTX 3060 12 GB
10.6 GB usable · 360 GB/s
3.3 GB 121 Runs great
M4 · 16 GB
10.7 GB usable · 120 GB/s
3.3 GB 37 Runs great
GeForce RTX 4090
22.4 GB usable · 1008 GB/s
3.3 GB 338 Runs great
Radeon RX 7900 XTX
22.4 GB usable · 960 GB/s
3.3 GB 322 Runs great
M3 Max · 64 GB
48.0 GB usable · 400 GB/s
3.3 GB 124 Runs great

All 114 devices, by tier →