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

distilbert-base-uncased-finetuned-sst-2-english

by distilbert

67M params · text-classification · 885 likes · 3.4M downloads

distilbert-base-uncased-finetuned-sst-2-english is a 67M parameter model. At Q4 quantization it requires 0GB of VRAM. It runs comfortably on GeForce RTX 4090 (19610 tok/s), GeForce RTX 5090 (29407 tok/s), M4 Max 128GB (7176 tok/s).

Inference providers

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

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB19610 tok/scomfortable
GeForce RTX 509032GB29407 tok/scomfortable
M4 Max 128GB128GB7176 tok/scomfortable
M4 Pro 48GB48GB3588 tok/scomfortable
M4 Pro 24GB24GB3588 tok/scomfortable
A100 PCIe 80 GB80GB35928 tok/scomfortable
H100 SXM5 80 GB80GB72263 tok/scomfortable
GeForce RTX 309024GB17338 tok/scomfortable
Radeon RX 7900 XTX24GB14337 tok/scomfortable
GeForce RTX 408016GB13917 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0