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

WizardCoder-15B-V1.0

by WizardLMTeam

15.0B params · text-generation · 763 likes · 395 downloads

WizardCoder-15B-V1.0 is a 15.0B parameter model. At Q4 quantization it requires 8GB of VRAM. It runs comfortably on GeForce RTX 4090 (87 tok/s), GeForce RTX 5090 (131 tok/s), M4 Max 128GB (32 tok/s).

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB87 tok/scomfortable
GeForce RTX 509032GB131 tok/scomfortable
M4 Max 128GB128GB32 tok/scomfortable
M4 Pro 48GB48GB16 tok/stight
M4 Pro 24GB24GB16 tok/stight
A100 PCIe 80 GB80GB160 tok/scomfortable
H100 SXM5 80 GB80GB322 tok/scomfortable
GeForce RTX 309024GB77 tok/scomfortable
Radeon RX 7900 XTX24GB64 tok/scomfortable
GeForce RTX 408016GB62 tok/scomfortable
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0