dexiio

Best local LLM for coding on a CPU only / integrated graphics (2026)

All figures assume an f16 KV cache, a 0.6 GB display reserve on the GPU, and 64 GB of DDR5 system RAM for the offload tiers. Tune these in the calculator.

The verdict

Qwen3 Coder 30B A3B Q8_0

CPU only at 32K · Coding score 9/10 · ≈ 14 tok/s · needs 38 GB system RAM

Purpose-built agentic coder. Best local fill-in-the-middle and tool-calling under 70B; useless at small talk.

llama-server -m Qwen3-Coder-30B-A3B-Instruct-Q8_0.gguf -c 32768 --flash-attn -ngl 0

Worthy alternates

Qwen3.5 27B Q8_0

CPU only · ≈ 2 tok/s · Coding 8/10

The dense 24GB workhorse. If you want one model on a 3090 and no surprises, it's this.

Qwen3.5 35B A3B Q8_0

CPU only · ≈ 13 tok/s · Coding 8/10

The meta pick, full stop. Near-dense-30B quality at 3B-active speed, and expert offload puts it on 8 GB cards.

Tune this for your exact RAM and settings in the calculator → · All models on the CPU only / integrated graphics

Frequently asked questions

What is the best local LLM for coding on a CPU only / integrated graphics?

Qwen3 Coder 30B A3B at Q8_0 — it scores 9/10 for coding and runs as "CPU only" at 32K context on the CPU only / integrated graphics.

How much context do I need for coding?

We recommend 32K tokens for coding (minimum 16K). These picks are computed at 32K.

How fast will it run on a CPU only / integrated graphics?

Roughly 14 tokens/sec for Qwen3 Coder 30B A3B — usable for interactive use.

Do I need more than 0 GB of VRAM for coding?

No — the pick above needs 0 GB of VRAM plus 38 GB of system RAM at 32K.

What settings should I use?

Start with our command: llama-server -m Qwen3-Coder-30B-A3B-Instruct-Q8_0.gguf -c 32768 --flash-attn -ngl 0 — then tune context and KV quant in the fit calculator.