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

xlm-roberta-large

by FacebookAI

561M params · fill-mask · 498 likes · 6.2M downloads

xlm-roberta-large is a 561M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (2339 tok/s), GeForce RTX 5090 (3508 tok/s), M4 Max 128GB (856 tok/s).

Inference providers

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

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB2339 tok/scomfortable
GeForce RTX 509032GB3508 tok/scomfortable
M4 Max 128GB128GB856 tok/scomfortable
M4 Pro 48GB48GB428 tok/scomfortable
M4 Pro 24GB24GB428 tok/scomfortable
A100 PCIe 80 GB80GB4286 tok/scomfortable
H100 SXM5 80 GB80GB8621 tok/scomfortable
GeForce RTX 309024GB2068 tok/scomfortable
Radeon RX 7900 XTX24GB1710 tok/scomfortable
GeForce RTX 408016GB1660 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0