llamaperf
Submit

Raon-1

1 report

Raon-1 29B · 4B active at 3-bit generates a median of 232.0 tokens per second on the NVIDIA RTX 5070 Ti, the hardware it is most run on, from one run that kept part of the model in system RAM (community reports, as of 29 Oct 2026).

As of 29 Oct 2026, Raon-1 29B · 4B active at 3-bit on the hardware it is most run on, with the median of runs that kept part of the model in system RAM (one device, one request):

Raon-1 VRAM requirements by size and quant →

How does Raon-1 run on your hardware?

Pick your GPU or Mac, then read the speeds people reported for Raon-1 on it, or estimate memory fit and speed.

Buying hardware? Compare GPUs →

Free to use. No account needed. Memory estimates and community measurements are labelled separately.

Run Raon-1 yourself? Add your numbers.

Your coding agent can run a speed test on your machine and send it here with a free account. Or paste a result you've already posted.

Benchmark with your agentPaste a result

Reported performance for Raon-1

Filter this model’s reports by setup →
Thin page (1 of 3 reports needed for indexing). Add yours.
Tone: positiveNVIDIA hardware
generation:
232.0 tokens/s
quant:
IQ3_KS (GGUF)

Reported by the source; GPU count, offloading and concurrent requests can change these figures. Prompt-processing speed is shown only when the source reports it; generation speed cannot tell us time to first token. Check the full setup before comparing.

User reports Raon-1-mini-preview at 232 tok/s on an RTX 5070 Ti 16GB, up from 50 tok/s. Setup is ik_llama.cpp with an IQ3_KS GGUF, weights streamed from system RAM on a machine with a Ryzen 5 5600X and 48GB DDR4. The 232 tok/s is the combined result of stacked optimizations, with weights streaming at about 2/3 of the card's peak bandwidth.

Oct 29, 2026

Get a weekly email of new Raon-1 reports on any GPU.

Email me new reports

Reported configurations

Individual observations from this page, not expected speeds or a ranking. Select a model to open its report, with the source and offloading conditions. Reported context may be a limit rather than actual input length.

Model / hardwareQuant / engineReported contextGeneration
Raon-1 29B (4B active) mini-preview
NVIDIA RTX 5070 Ti
IQ3_KS
ik_llama.cpp
Not reported232.0 tokens/s

Frequently asked

What hardware do people run Raon-1 on?

The hardware Raon-1 29B · 4B active at 3-bit is most run on: NVIDIA RTX 5070 Ti (232.0 t/s median, one run). Medians of runs that kept part of the model in system RAM, as of 29 Oct 2026.

Which inference engines do people use for Raon-1?

Reports of Raon-1 name ik_llama.cpp (1 report), as of 29 Oct 2026.