What is the best LLM for a Mac with 8 GB?

Of the models with a published Intelligence Index, MiniCPM5 2B scores highest among those that run well in 8 GB (8-bit). The biggest that runs well is Granite 4.2 8B.

Is 8 GB enough for local AI on a Mac?

16 of the 73 open models here run well in it at 4 bits or better. macOS lets the GPU use about 6 GB of it by default; the model and its context have to fit in that.

Can a 8 GB Mac run gpt-oss-20b?

No: it needs more than the 6 GB the GPU can use by default.

Best local LLMs for a 8 GB Mac (M1 to M3): what runs, how fast

Macs with 8 GB

3 chips from M1 to M3 come with 8 GB. Open one for its own list and speeds.

Runs well 21

Fits with room for context, at 10 or more tokens a second on the M2.

Model	Version	Memory	M1	M2
MiniCPM5 2B2.52B · Intelligence 12.5	8-bit	2.7 GB	11–20	17–30
Granite 4.2 8B8B · Intelligence 11.1	4-bit	4.5 GB	7–12	10–18
Granite 4.2 3B3B · Intelligence 9.1	8-bit	3.2 GB	10–17	14–25
Gemma 4 E4B8B · Intelligence 8.9	4-bit	4.5 GB	12–22	18–32
Nemotron 3 Nano 4B3.97B · Intelligence 4.8	8-bit	4.2 GB	10–17	11–19
Llama 3.1 8B Instruct8B	4-bit	4.5 GB	7–12	10–18
Gemma 4 E2B5.1B	6-bit	4.1 GB	16–29	24–42
Qwen3.5 4B4B	8-bit	4.3 GB	13–24	10–19
Phi-4 mini instruct3.8B	8-bit	4 GB	10–18	11–20
Llama 3.2 3B Instruct3.21B	8-bit	3.4 GB	12–21	13–24
SmolLM3 3B3B	8-bit	3.2 GB	10–17	14–25
LFM2.5 2.6B2.69B	8-bit	2.9 GB	11–19	16–28
Qwen3.5 2B2B	8-bit	2.1 GB	15–26	21–38
Llama 3.2 1B Instruct1.23B	8-bit	1.3 GB	24–42	35–62
Qwen3.5 0.8B0.8B	8-bit	0.9 GB	34–60	50–89
Qwen3 0.6B0.6B	8-bit	0.6 GB	51–91	75–133
Qwen3.5 9B9B · Intelligence 11.2	3-bit lower quality	3.9 GB	8–14	12–21
Ministral 3 8B8.9B · Intelligence 5.5	3-bit lower quality	3.9 GB	8–14	12–21
Phi-4 (14B)14B	2-bit lower quality	4.4 GB	7–12	10–18
Gemma 4 12B Unified12B	2-bit lower quality	3.7 GB	8–15	12–22
Qwen3 8B8.2B	3-bit lower quality	3.6 GB	9–15	13–22

Only just fits 1

Needs almost all memory: raise the GPU memory limit and keep the context short.

Model	Version	Memory	M1	M2
Phi-4 Reasoning Vision 15B15B	2-bit lower quality	4.7 GB	7–12	10–17

Too big 51

Needs more memory than 8 GB, even at 2 bits.

Questions

What is the best LLM for a Mac with 8 GB?: Of the models with a published Intelligence Index, MiniCPM5 2B scores highest among those that run well in 8 GB (8-bit). The biggest that runs well is Granite 4.2 8B.
Is 8 GB enough for local AI on a Mac?: 16 of the 73 open models here run well in it at 4 bits or better. macOS lets the GPU use about 6 GB of it by default; the model and its context have to fit in that.
Can a 8 GB Mac run gpt-oss-20b?: No: it needs more than the 6 GB the GPU can use by default.

Other sizes: 8 GB · 16 GB · 18 GB · 24 GB · 32 GB · 36 GB · 48 GB · 64 GB · 96 GB · 128 GB · 192 GB · 256 GB · 512 GB

Run them with MLX Harness (free, open source) or Apple’s mlx-lm in Terminal: step by step.

How it’s worked out

Memory: the model’s parameters at 8, 6, 4, 3 or 2 bits per weight plus MLX’s group scales (half a bit per weight), or its native format, with room for the context on top. By default macOS lets the GPU use about 75% of unified memory. Speed: memory bandwidth divided by the bytes read per token is the ceiling for one conversation; in llama.cpp’s public Apple Silicon benchmarks real runs reach about 45% to 80% of it, which is the range shown. These are estimates, not measurements: measure your own Mac on MLX Bench. Model figures from each official model card, chip figures from Apple’s Tech Specs. Checked 2026-10-19. Same data as mlx.app/compare.