Auto-tuning & speed

TurboLLM auto-tunes to your hardware on load: it benchmarks your exact GPU, derives fast launch defaults, and shows a VRAM-fit verdict before you load — so there's no flag guessing. It also reports real, measured tokens/sec, never a faked number.

What it tunes

On load, TurboLLM derives fast defaults for the flags that actually move the needle, so you don't have to hand-tune them per model:

Context & offload

Context length (-c), GPU offload (-ngl), and MoE CPU-offload (--n-cpu-moe).

Cache & throughput

KV-cache quant type, parallel slots, and CPU threads.

Speed features

Flash attention and speculative decoding — NextN self-speculative for models that carry a draft head.

Fast by default

TurboLLM turns on the speed wins automatically:

For sampling, auto-tune uses the recommended settings from Hugging Face: it checks a repo's structured params / generation_config.json sidecar first, so defaults match what the model author intended.

VRAM-fit verdict

Before you commit to a load, TurboLLM shows whether the configuration fits your VRAM — so you learn a model won't fit before waiting on a load, not after it fails.

VRAM headroom slider

Under Settings → Engine, the headroom slider (300 MB–2 GB, default 1 GB) tells auto-tune how much VRAM to keep free for other GPU workloads — ComfyUI, browser tabs, and anything else sharing the card.

Auto-fit GPU layers / MoE CPU offload

New in v1.7.7

Two toggles let llama.cpp decide the GPU/CPU split for you at load time instead of using a fixed number — handy when a large context or model doesn't fit a fixed default. Both are off by default.

Real measured tokens/sec

The model list shows genuine throughput, not a marketing figure:

Profiles are saved per model, per engine, and you can run a model multi-GPU — splitting it across cards.

Speed comparison

Same GPU, same model, same 200K context — measured generation

RTX 5070 Ti 16 GB · Qwen3.6-35B-A3B at 200K context. Every number below is measured generation, not estimated:

BuildTurboLLMLM StudioSpeed-up
official llama.cpp — q4_074.7 t/s61.0 t/s1.2×
official llama.cpp — q8_072.3 t/s~66 t/s1.1×
TurboQuant fork — turbo424.6 t/s11.4 t/s2.2×