Best local LLM for summarization 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
Qwen3.5 27B Q8_0
CPU only at 32K · Summarization score 8/10 · ≈ 2 tok/s · needs 39 GB system RAM
The dense 24GB workhorse. If you want one model on a 3090 and no surprises, it's this.
llama-server -m Qwen3.5-27B-Q8_0.gguf -c 32768 --flash-attn -ngl 0
Worthy alternates
Qwen3.5 35B A3B Q8_0
CPU only · ≈ 13 tok/s · Summarization 8/10
The meta pick, full stop. Near-dense-30B quality at 3B-active speed, and expert offload puts it on 8 GB cards.
Gemma 3 27B Q8_0
CPU only · ≈ 2 tok/s · Summarization 8/10
Sliding-window attention keeps long context cheap, and the vision stack still beats most of 2026's newcomers.
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 summarization on a CPU only / integrated graphics?
Qwen3.5 27B at Q8_0 — it scores 8/10 for summarization and runs as "CPU only" at 32K context on the CPU only / integrated graphics.
How much context do I need for summarization?
We recommend 48K tokens for summarization (minimum 24K). These picks are computed at 32K.
How fast will it run on a CPU only / integrated graphics?
Roughly 2 tokens/sec for Qwen3.5 27B — usable for interactive use.
Do I need more than 0 GB of VRAM for summarization?
No — the pick above needs 0 GB of VRAM plus 39 GB of system RAM at 32K.
What settings should I use?
Start with our command: llama-server -m Qwen3.5-27B-Q8_0.gguf -c 32768 --flash-attn -ngl 0 — then tune context and KV quant in the fit calculator.