Can RTX 4090 run gpt-oss 20B?

Q: Can RTX 4090 run gpt-oss 20B?

Yes. Measured 279.65 tok/s generation on RTX 4090, via llama.cpp. Source: https://github.com/geerlingguy/ai-benchmarks/issues/43

Yes. Measured 279.65 tok/s generation (llama.cpp; source).

Every measured result for gpt-oss 20B on RTX 4090

Build	Quant / setup	Engine	tok/s
Intel Core Ultra 265K + RTX 4090		llama.cpp	279.65	source
RTX 4090: 2026 models (llamabench community)	llama-bench tg128 (quant per result)	llama.cpp	203	source

Memory: RTX 4090 has 24 GB. gpt-oss 20B weights are about 13 GB in its native format (GGUF param count via llamabench.ai (OpenAI model card: 21B total, 3.6B active)). Arithmetic, not a measurement; long contexts need more. Check other quants in the VRAM calculator.

Speeds are decode (generation) tokens/sec as reported by each source; quant, engine and context differ, so small gaps are noise.

All measured speeds for gpt-oss 20B: best home rig for gpt-oss 20B.

gpt-oss 20B on other hardware: Apple M1 Max · Apple M1 Ultra · Apple M2 Ultra · Apple M3 Ultra · Apple M4 Max · DGX Spark (GB10) · RTX 3060 12GB · RTX 3090 · RTX 4070 Ti · RTX 5060 Ti 16GB · RTX 5070 Ti · RTX 5080 · RTX 5090 · RTX A4000 · RTX Pro 6000 (Blackwell, 96 GB) · RTX Spark N1X · Radeon AI Pro R9700

Other models measured on RTX 4090: Flint 40B-A2.5B · GLM 4.7 Flash · Gemma 4 12B · Gemma 4 26B-A4B · Gemma 4 31B · LFM2.5 24B-A2B · Nemotron 3.5 Bolt 74B-A9B · Qwen3.5 9B · Qwen3.6 27B · Qwen3.6 35B-A3B

Not enough VRAM? Rent before you buy. Run gpt-oss 20B on a bigger GPU on a rented 24–96 GB cloud GPU by the hour and see the real speed before spending €1,000+. Try it on Runpod
Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms). It never affects our data.