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

deepseek-coder-33b-instruct

by deepseek-ai

33.3B params · text-generation · 565 likes · 10.4k downloads

deepseek-coder-33b-instruct is a 33.3B parameter model. At Q4 quantization it requires 17GB of VRAM. It runs comfortably on GeForce RTX 4090 (39 tok/s), GeForce RTX 5090 (59 tok/s), A100 PCIe 80 GB (72 tok/s).

GPU compatibility

GPUVRAMQ4 DecodeVerdict
GeForce RTX 409024GB39 tok/scomfortable
GeForce RTX 509032GB59 tok/scomfortable
M4 Max 128GB128GB14 tok/stight
M4 Pro 48GB48GB7 tok/stight
M4 Pro 24GB24GB7 tok/stight
A100 PCIe 80 GB80GB72 tok/scomfortable
H100 SXM5 80 GB80GB145 tok/scomfortable
GeForce RTX 309024GB34 tok/scomfortable
Radeon RX 7900 XTX24GB28 tok/stight
M4 Max 64GB64GB14 tok/stight
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0