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

cryptobert

by ElKulako

125M params · text-classification · 185 likes · 136.2k downloads

cryptobert is a 125M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (10533 tok/s), GeForce RTX 5090 (15796 tok/s), M4 Max 128GB (3854 tok/s).

Inference providers

Provider$/1M in$/1M outThroughput
HF Inference

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB10533 tok/scomfortable
GeForce RTX 509032GB15796 tok/scomfortable
M4 Max 128GB128GB3854 tok/scomfortable
M4 Pro 48GB48GB1927 tok/scomfortable
M4 Pro 24GB24GB1927 tok/scomfortable
A100 PCIe 80 GB80GB19299 tok/scomfortable
H100 SXM5 80 GB80GB38816 tok/scomfortable
GeForce RTX 309024GB9313 tok/scomfortable
Radeon RX 7900 XTX24GB7701 tok/scomfortable
GeForce RTX 408016GB7475 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0