Llama 3.2 Vision on Apple M3 Pro · 18GB

✅ Runs (tight) — Llama 3.2 Vision 11B @ Q4_K_M
Llama 3.2 Vision 11B at Q4_K_M needs ~7.8 GB (weights 6.0 GB + KV 1.3 GB + overhead 512 MB @ 8K ctx) of your 9.1 GB usable (of 18.0 GB unified memory) — fits, but little headroom — close other apps or trim context. Expect ~11–18 tok/s (usable).
context @ Q4_K_M: comfortable to 4K · runs to 32K · won't fit past that on this rig
quantneedsspeed
~ Q8_012.3 GB~0.8–1 tok/s (slow)
✓ Q4_K_M7.8 GB~11–18 tok/s (usable)
$runlocal install llama3.2-vision:11b
runlocal verdict card
download card share on x share on reddit drop it in your model-drop post
Llama 3.2 Vision: Runs (tight)
for your own README — links back here

Same model, other machines: Apple M1 · Apple M1 Pro · Apple M2 · Apple M2 Pro · Apple M4 Pro · Apple M3 Max

Also runs on a Apple M3 Pro: DeepSeek-R1 (distill) · Qwen2.5 · Mistral 7B · Qwen3 · Llama 3.1

Best models for a Apple M3 Pro · 18GB →

Architecturally similar

These share Llama 3.2 Vision's KV-cache geometry — they size memory the same way, so here's how they fit on a Apple M3 Pro · 18GB too:

modelsharesverdict here
Qwen2.5-Coder 8kv/128hd ✅ Runs (tight)
Llama 3.2 8kv/128hd ✅ Runs comfortably
Llama 3.3 8kv/128hd ❌ Won't fit
Mistral Small 3 8kv/128hd 🐢 CPU-only (slow)
Mistral Nemo 8kv/128hd ✅ Runs (tight)
Codestral 8kv/128hd 🐢 CPU-only (slow)
Mixtral 8x7B 8kv/128hd ❌ Won't fit
Phi-4-mini 8kv/128hd ✅ Runs comfortably

…and 6 more — see the Llama 3.2 Vision page.

Check any combo yourself: open the checker →