rnj-1-instruct

RNJ family

1recorded run

123.2best tok/s

0verified

Evidence: 0 sourced runs · 1 awaiting review · 1 independent source · first recorded Aug 17, 2026. What these labels mean.

What the rnj-1-instruct numbers mean

The fastest result on our board is 123.2 output tok/s on an RTX 5090 at Q8_0 with a 2048-token context. That is about 8.1 ms per generated token — 500 tokens takes roughly 4.1 seconds. Prompt processing on the same record runs at 1673.1 tok/s, so a 2,000-token prompt is read in about 1.2 seconds.

Fastest GPUs for this model (best tok/s)

RTX 5090123.2

All benchmarks Methodology

GPU	Model	Quant	Backend	Ctx	tok/s	prompt tok/s	VRAM GB	Source
RTX 5090 · 32 GB	rnj-1-instruct	Q8_0	lmstudio —	2048	123.2 …	1673.1	—	src