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

bge-large-en

by BAAI

335M params · feature-extraction · 223 likes · 207.0k downloads

bge-large-en is a 335M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (3917 tok/s), GeForce RTX 5090 (5875 tok/s), M4 Max 128GB (1433 tok/s).

Inference providers

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

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB3917 tok/scomfortable
GeForce RTX 509032GB5875 tok/scomfortable
M4 Max 128GB128GB1433 tok/scomfortable
M4 Pro 48GB48GB716 tok/scomfortable
M4 Pro 24GB24GB716 tok/scomfortable
A100 PCIe 80 GB80GB7177 tok/scomfortable
H100 SXM5 80 GB80GB14436 tok/scomfortable
GeForce RTX 309024GB3463 tok/scomfortable
Radeon RX 7900 XTX24GB2864 tok/scomfortable
GeForce RTX 408016GB2780 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0