Can an NVIDIA GeForce RTX 5090 run gpt-oss-20b?
Yes. gpt-oss-20b needs 13.0GB at MXFP4 with llama.cpp and 8K context; the NVIDIA GeForce RTX 5090 has 32GB of VRAM (31.5GB usable) — grade A+, with decode speed in the interactive tier.
NVIDIA GeForce RTX 5090 · gpt-oss-20b · MXFP4
✅ Fits — grade A+
InteractivebetaHybrid attention — KV computed as full-cache upper bound · MoE speed estimate — wide error bars
weights 12.1 GBKV cache 0.40 GBoverhead 0.44 GBpool 31.5 GBneeds 13.0 GB
assumes llama.cpp · batch 1 · 8K context · fp16 KV cache
$1,999 MSRP · as of 2026-07-08
Affiliate link — we may earn a commission. Verdicts are computed before any link is attached.
Every quant, graded
NVIDIA GeForce RTX 5090 × gpt-oss-20b, llama.cpp, 8K context.
| Quant | File size | Fit | Speed tier |
|---|---|---|---|
| MXFP4 | 12.1 GB | Fits · A+ | interactive · beta |
gpt-oss-20b on similar hardware
More models on the NVIDIA GeForce RTX 5090
Verdict computed from published specs and measured GGUF file sizes — see how the math works.