GPUs & Macs › NVIDIA RTX 5080
NVIDIA RTX 5080: local LLM speed
Llama 2 7B (Q4_0) generates 192 tokens/s on the NVIDIA RTX 5080 (Vulkan), and processes prompts at 8,055 tokens/s.
Measured results
| Model | Generate tok/s | Prompt tok/s | Quant | Backend | Test | Source |
|---|---|---|---|---|---|---|
| Llama 2 7B | 192 | 8,055 | Q4_0 | Vulkan FA on | tg128 | llama.cpp #10879 details |
| Llama 2 7B | 185 | 7,445 | Q4_0 | Vulkan FA off | tg128 | llama.cpp #10879 details |
| Llama 2 7B | 185 | 9,488 | Q4_0 | CUDA FA on | tg128 | llama.cpp #15013 details |
| Llama 2 7B | 182 | 8,297 | Q4_0 | CUDA FA off | tg128 | llama.cpp #15013 details |
| gpt-oss 20B | 229 | 10,404 | MXFP4 | CUDA FA on | tg128 · n=4 | llama.cpp #15396 llama.cpp #15396 details |
| gpt-oss 20B | 178 | 12,995 | F16 | CUDA FA on | tg128 | llama.cpp #15396 details |
| Llama 3.2 1B Instruct | 140 | 13,261 | Q4_K_M | llamafile | LocalScore avg · n=2 | LocalScore LocalScore details |
| Llama 3.1 8B Instruct | 35.5 | 4,045 | Q4_K_M | llamafile | LocalScore avg · n=2 | LocalScore LocalScore details |
| Qwen2.5 14B Instruct | 29.3 | 2,256 | Q4_K_M | llamafile | LocalScore avg · n=2 | LocalScore LocalScore details |
What fits on the NVIDIA RTX 5080
Estimates for this chip are scaled from a measured run: Llama 2 7B Q4_0 at 192 tok/s (Vulkan, source). Fit assumes the whole model in GPU memory with ~8K context.
| Model | Best quant that fits | File | Expected tok/s |
|---|---|---|---|
| Qwen3.8-27B | UD-IQ3_XXS | 10.9 GB | 67.2 estimate |
| gemma-4-12B-it | Q8_0 | 12.7 GB | 58.0 estimate |
| Qwen3.5-9B | Q8_0 | 9.5 GB | 77.1 estimate |
| Qwen3.5-4B | Q8_0 | 4.5 GB | 164 estimate |
| Qwen3.6-35B-A3B | UD-IQ3_XXS | 13.2 GB | 324 estimate |
| gemma-4-26B-A4B-it | UD-IQ4_XS | 13.6 GB | 176 estimate |
| MiMo-V2.6-Distill-Qwen-9B | Q8_0 | 9.5 GB | 77.0 estimate |
| gemma-4-E4B-it | Q8_0 | 8.2 GB | 89.7 estimate |
| Muse-Glimmer-30B | UD-IQ3_XXS | 13.1 GB | 55.9 estimate |
| Qwen3-4B | Q8_0 | 4.3 GB | 172 estimate |
| gemma-4-31B-it | UD-IQ3_XXS | 11.8 GB | 62.1 estimate |
| gemma-4-12B-it-qat-q4_0-unquantized | UD-Q4_K_XL | 6.7 GB | 109 estimate |
| gemma-3-1b-it | Q8_0 | 1.1 GB | 612 estimate |
| gemma-4-E2B-it | Q8_0 | 5.0 GB | 148 estimate |
Similar hardware
NVIDIA RTX 5070 Ti · NVIDIA RTX 4080 SUPER · NVIDIA RTX 4080 · NVIDIA RTX 4070 Ti SUPER · NVIDIA RTX 5060 Ti · NVIDIA RTX PRO 2000 Blackwell · NVIDIA Quadro P5000 · NVIDIA Quadro RTX 5000