Code Llama on NVIDIA GeForce RTX 4090 · 24GB VRAM

✅ Runs (tight) — Code Llama 34B @ Q4_K_M
Code Llama 34B at Q4_K_M needs ~21.5 GB (weights 19.1 GB + KV 1.5 GB + overhead 975 MB @ 8K ctx) of your 21.6 GB usable VRAM — fits, but little headroom — close other apps or trim context. Expect ~19–30 tok/s (fast).
context @ Q4_K_M: runs to its full 16K
quantneedsspeed
✗ Q8_036.6 GB—
~ Q5_K_M25.0 GB~5–9 tok/s (slow)
✓ Q4_K_M21.5 GB~19–30 tok/s (fast)
✓ Q3_K_M17.6 GB~25–40 tok/s (fast)
$runlocal install codellama:34b
⚠ 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
Code Llama: 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 4090: DeepSeek-R1 (distill) · Qwen3 · Qwen2.5 · Gemma 3 · Gemma 4

Best models for a NVIDIA GeForce RTX 4090 · 24GB VRAM →

Architecturally similar

These share Code Llama's KV-cache geometry — they size memory the same way, so here's how they fit on a NVIDIA GeForce RTX 4090 · 24GB VRAM too:

modelsharesverdict here
Llama 3.1 8kv/128hd ✅ Runs comfortably
Qwen2.5-Coder 8kv/128hd ✅ Runs (tight)
Llama 3.2 8kv/128hd ✅ Runs comfortably
Llama 3.3 8kv/128hd ⚠️ Partial GPU offload
Mistral 7B 8kv/128hd ✅ Runs comfortably
Mistral Small 3 8kv/128hd ✅ Runs (tight)
Mistral Nemo 8kv/128hd ✅ Runs comfortably
Codestral 8kv/128hd ✅ Runs comfortably

…and 10 more — see the Code Llama page.

Check any combo yourself: open the checker →