vram.run Models Hardware Providers Cloud State of Inference
API provider data is live · Hardware & cloud pricing curated 2026-02-23

Broken-Tutu-24B-Unslop-v2.0

by ReadyArt

23.6B params · text-generation · 85 likes · 586 downloads

Broken-Tutu-24B-Unslop-v2.0 is a 23.6B parameter model. At Q4 quantization it requires 12GB of VRAM. It runs comfortably on GeForce RTX 4090 (55 tok/s), GeForce RTX 5090 (83 tok/s), A100 PCIe 80 GB (102 tok/s).

Inference providers

Provider$/1M in$/1M outThroughput
Featherless

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB55 tok/scomfortable
GeForce RTX 509032GB83 tok/scomfortable
M4 Max 128GB128GB20 tok/stight
M4 Pro 48GB48GB10 tok/stight
M4 Pro 24GB24GB10 tok/stight
A100 PCIe 80 GB80GB102 tok/scomfortable
H100 SXM5 80 GB80GB205 tok/scomfortable
GeForce RTX 309024GB49 tok/scomfortable
Radeon RX 7900 XTX24GB40 tok/scomfortable
GeForce RTX 408016GB39 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0