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

bert-large-uncased

by google-bert

336M params · fill-mask · 145 likes · 635.8k downloads

bert-large-uncased is a 336M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (3905 tok/s), GeForce RTX 5090 (5856 tok/s), M4 Max 128GB (1429 tok/s).

Inference providers

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

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB3905 tok/scomfortable
GeForce RTX 509032GB5856 tok/scomfortable
M4 Max 128GB128GB1429 tok/scomfortable
M4 Pro 48GB48GB714 tok/scomfortable
M4 Pro 24GB24GB714 tok/scomfortable
A100 PCIe 80 GB80GB7154 tok/scomfortable
H100 SXM5 80 GB80GB14390 tok/scomfortable
GeForce RTX 309024GB3452 tok/scomfortable
Radeon RX 7900 XTX24GB2855 tok/scomfortable
GeForce RTX 408016GB2771 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0