Can the NVIDIA RTX 4080 run gpt-oss 20B?
Yes
QuantQ5_K_MVery good
Memory15.9 GBof 16 GB
Speed~168tokens/s
Yes. The NVIDIA RTX 4080 can run gpt-oss 20B fully in VRAM at Q5_K_M (very good), using about 15.9 GB of the 16 GB available, at roughly 168 tokens/second.
Try other context lengths in the calculator →Memory needed by quantization
At 8,192 tokens of context.
| Quant | Quality | Memory | Fits? | Speed |
|---|---|---|---|---|
| FP16 | Full quality | 42.8 GB | Too big | — |
| Q8_0 | Near-lossless | 23.2 GB | Too big | — |
| Q6_K | Excellent | 18.1 GB | Too big | — |
| Q5_K_M | Very good | 15.9 GB | Fits | ~168 tok/s |
| Q4_K_M | Good, most popular | 13.7 GB | Fits | ~197 tok/s |
| Q3_K_M | Noticeably degraded | 11.2 GB | Fits | ~245 tok/s |
| Q2_K | Heavily degraded | 8.8 GB | Fits | ~319 tok/s |
How to run gpt-oss 20B on the NVIDIA RTX 4080
With Ollama installed, one command downloads and starts it:
ollama run gpt-oss:20bOllama's default download is Q4_K_M (about 13.7 GB here). You have room for Q5_K_M, which is very good: pick that GGUF in LM Studio or from Hugging Face.
GPUs that run gpt-oss 20B comfortably
- NVIDIA RTX 4060 Ti 16GB
- NVIDIA RTX 5060 Ti 16GB
- AMD Radeon RX 7800 XT (16 GB)
- NVIDIA RTX 4070 Ti Super (16 GB)
- NVIDIA RTX 5070 Ti (16 GB)
Models that run well on the NVIDIA RTX 4080
Run it in the cloud
Don't want to buy hardware? Run gpt-oss 20B on a rented GPU today.
Rent a cloud GPU by the hour. No upfront cost, cancel any time.