Yes — with 2.5 GB to spare
Gemma 4 12B at Q4_K_M fits your M1 · 16 GB entirely in unified memory at 8K context, at an estimated 5.6 tokens per second. Past 48K 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 · 6.7 GB
Apache 2.0
Released 29 May 2026
Vision
The "unified" Gemma 4: text, image and audio in one 12B that fits a 12 GB card at Q4. 140+ languages.
The VRAM budget
weights 6.7 GB
Weights 6.7 GB
KV cache @ 8K 0.81 GB
Runtime overhead 0.6 GB
Free 2.5 GB of 10.7 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 22.4 GB | 23.8 GB | — | ~1.7 | Reference | 13.1 GB over |
| Q8_0 | 11.9 GB | 13.3 GB | — | ~3.2 | −0.1% ppl | 2.6 GB over |
| Q6_K | 9.2 GB | 10.6 GB | 9K | 4.1 | −0.4% ppl | Fits |
| Q5_K_M | 7.9 GB | 9.3 GB | 29K | 4.8 | −0.8% ppl | Long context |
| Q4_K_M | 6.7 GB | 8.2 GB | 48K | 5.6 | −1.9% ppl | Recommended |
| Q3_K_M | 5.5 GB | 6.9 GB | 69K | 7.0 | −5.4% ppl | Long context |
| Q2_K | 4.7 GB | 6.1 GB | 81K | 8.1 | −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. This model interleaves sliding-window layers (1024 tokens, 8 of 48 layers global), which is why its cache barely grows with context.
How to run it
$ pip install mlx-lm $ mlx_lm.generate --model mlx-community/gemma-4-12B-it-4bit \ --max-tokens 512 --prompt "Hello"
Apple's own array framework. The fastest path on Apple Silicon. More on MLX.
01Download is 6.7 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 10.7 GB of 16 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.
03There is room to go to 48K context on this card.