LLaVA on NVIDIA GeForce RTX 5080 · 16GB VRAM

✅ Runs (tight) — LLaVA 13B @ Q4_K_M
LLaVA 13B at Q4_K_M needs ~14.3 GB (weights 7.5 GB + KV 6.3 GB + overhead 512 MB @ 8K ctx) of your 14.4 GB usable VRAM — fits, but little headroom — close other apps or trim context. Expect ~25–45 tok/s (fast).
context @ Q4_K_M: comfortable to 4K
quantneedsspeed
~ Q8_020.1 GB~6–10 tok/s (slow)
~ Q5_K_M15.6 GB~8–13 tok/s (usable)
✓ Q4_K_M14.3 GB~25–45 tok/s (fast)
$runlocal install llava:13b
⚠ NVIDIA support is best-effort in v0.1 — verify before relying on it.
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 M2 Pro · Apple M4 Pro

Also runs on a NVIDIA GeForce RTX 5080: DeepSeek-R1 (distill) · Qwen2.5 · Qwen3 · Gemma 4 · Gemma 3

Best models for a NVIDIA GeForce RTX 5080 · 16GB VRAM →

Architecturally similar

These share LLaVA's KV-cache geometry — they size memory the same way, so here's how they fit on a NVIDIA GeForce RTX 5080 · 16GB VRAM too:

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

Check any combo yourself: open the checker →