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

medgemma-27b-text-it

by google

27.0B params · text-generation · 444 likes · 37.1k downloads

medgemma-27b-text-it is a 27.0B parameter model. At Q4 quantization it requires 14GB of VRAM. It runs comfortably on GeForce RTX 4090 (48 tok/s), GeForce RTX 5090 (72 tok/s), A100 PCIe 80 GB (89 tok/s).

Inference providers

Provider$/1M in$/1M outThroughput
Featherless

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB48 tok/scomfortable
GeForce RTX 509032GB72 tok/scomfortable
M4 Max 128GB128GB17 tok/stight
M4 Pro 48GB48GB8 tok/stight
M4 Pro 24GB24GB8 tok/stight
A100 PCIe 80 GB80GB89 tok/scomfortable
H100 SXM5 80 GB80GB179 tok/scomfortable
GeForce RTX 309024GB42 tok/scomfortable
Radeon RX 7900 XTX24GB35 tok/scomfortable
GeForce RTX 408016GB34 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0