Can the RTX 5080 run gpt-oss-20b?
Just. gpt-oss-20b at Q4_K_M needs about 13.8 GB of the RTX 5080's 16 GB, running at an estimated 183 tok/s. Keep context short and close other GPU apps.
Every quantization
gpt-oss-20b (21.5B, 3.6B active) on 16 GB
| Quant | Weights | Needs | Fit | Est. speed |
|---|---|---|---|---|
| Q4_K_M | 12 GB | 13.8 GB | Tight | ~183 tok/s |
| Q5_K_M | 15 GB | 16.8 GB | Too big | - |
| Q8_0 | 23 GB | 24.8 GB | Too big | - |
| FP16 | 43 GB | 44.8 GB | Too big | - |
Estimates. Needs = weights + 1.8 GB for KV cache and runtime at modest context. Speed is single-stream decode scaled from the RTX 5080's 960 GB/s bandwidth (Estimated from memory bandwidth (llama.cpp decode)). Long contexts need more memory and run slower.
Run it
Copy, paste, go
llama-server -hf unsloth/gpt-oss-20b-GGUF -ngl 99Ollama pulls a Q4 build by default. Pick a larger quant only if the table above says it fits. Full gpt-oss-20b details →
Compare
Same model, other hardware - and other models on this card
gpt-oss-20b on other hardware
- RTX 5090~341 tok/s
- RTX 5090 D~341 tok/s
- RTX 6000D~299 tok/s
- RTX 4090~208 tok/s
- Radeon RX 7900 XTX~183 tok/s
- RTX 3090~182 tok/s
- RTX 5080 Mobile~171 tok/s
- MacBook Pro 16" M4 Max~104 tok/s
Also runs on the RTX 5080
- Qwen2.5-Coder-14B-Instruct~61 tok/s
- Qwen3 14B~69 tok/s
- Phi-4~65 tok/s
- Gemma 4 12B~79 tok/s
- Qwen3.5 9B~92 tok/s
- Llama 3.1 8B~100 tok/s
- Mistral-7B-Instruct-v0.2~138 tok/s
- Qwen3.5 4B~183 tok/s
Frequently asked
Quick answers
Can the NVIDIA RTX 5080 run gpt-oss-20b?
Just. gpt-oss-20b at Q4_K_M needs about 13.8 GB of the RTX 5080's 16 GB, running at an estimated 183 tok/s. Keep context short and close other GPU apps.
How much VRAM does gpt-oss-20b need?
About 12 GB of weights at Q4_K_M, 23 GB at Q8_0 and 43 GB at FP16. Add roughly 1.8 GB for the KV cache and runtime at modest context lengths.
How fast is gpt-oss-20b on the RTX 5080?
Roughly 183 tokens per second at Q4_K_M (single user, llama.cpp-style decode). This is an estimate from the RTX 5080's 960 GB/s memory bandwidth; as a mixture-of-experts model only about 3.6B parameters are read per token, which is why it is faster than its size suggests.