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

mDeBERTa-v3-base-mnli-xnli

by MoritzLaurer

279M params · zero-shot-classification · 296 likes · 144.4k downloads

mDeBERTa-v3-base-mnli-xnli is a 279M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (4709 tok/s), GeForce RTX 5090 (7062 tok/s), M4 Max 128GB (1723 tok/s).

Inference providers

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

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB4709 tok/scomfortable
GeForce RTX 509032GB7062 tok/scomfortable
M4 Max 128GB128GB1723 tok/scomfortable
M4 Pro 48GB48GB861 tok/scomfortable
M4 Pro 24GB24GB861 tok/scomfortable
A100 PCIe 80 GB80GB8628 tok/scomfortable
H100 SXM5 80 GB80GB17353 tok/scomfortable
GeForce RTX 309024GB4163 tok/scomfortable
Radeon RX 7900 XTX24GB3443 tok/scomfortable
GeForce RTX 408016GB3342 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0