LLaVA on Apple M2 Pro · 32GB

✅ Runs (tight) — LLaVA 13B @ Q5_K_M
LLaVA 13B at Q5_K_M needs ~15.6 GB (weights 8.8 GB + KV 6.3 GB + overhead 512 MB @ 8K ctx) of your 18.4 GB usable (of 32.0 GB unified memory) — fits, but little headroom — close other apps or trim context. Expect ~7–11 tok/s (usable).
context @ Q5_K_M: comfortable to 4K
quantneedsspeed
~ Q8_020.1 GB~0.6–1 tok/s (slow)
✓ Q5_K_M15.6 GB~7–11 tok/s (usable)
✓ Q4_K_M14.3 GB~8–13 tok/s (usable)
$runlocal install llava:13b
runlocal verdict card
download card share on x share on reddit drop it in your model-drop post
LLaVA: Runs (tight)
for your own README — links back here

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

Also runs on a Apple M2 Pro: DeepSeek-R1 (distill) · Mistral Small 3 · Qwen3 · Llama 3.1 · Qwen2.5

Best models for a Apple M2 Pro · 32GB →

Architecturally similar

These share LLaVA's KV-cache geometry — they size memory the same way, so here's how they fit on a Apple M2 Pro · 32GB too:

modelsharesverdict here
Code Llama 32kv/128hd40kv/128hd ✅ Runs (tight)
OLMo 2 7B 32kv/128hd ✅ Runs comfortably

Check any combo yourself: open the checker →