GPUs & Macs › NVIDIA RTX 3500 Ada Laptop
NVIDIA RTX 3500 Ada Laptop: local LLM speed
3 measured results for the NVIDIA RTX 3500 Ada Laptop.
Measured results
| Model | Generate tok/s | Prompt tok/s | Quant | Backend | Test | Source |
|---|---|---|---|---|---|---|
| Llama 3.2 1B Instruct | 181 | 8,206 | Q4_K_M | llamafile | LocalScore avg | LocalScore details |
| Llama 3.1 8B Instruct | 41.8 | 2,445 | Q4_K_M | llamafile | LocalScore avg | LocalScore details |
| Qwen2.5 14B Instruct | 25.1 | 1,298 | Q4_K_M | llamafile | LocalScore avg | LocalScore details |
What fits on the NVIDIA RTX 3500 Ada Laptop
Estimates for this chip are scaled from a measured run: Qwen2.5 14B Instruct Q4_K_M at 25.1 tok/s (llamafile, 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-IQ2_XXS | 7.3 GB | 31.0 estimate |
| gemma-4-12B-it | Q6_K | 9.8 GB | 23.0 estimate |
| Qwen3.5-9B | Q8_0 | 9.5 GB | 23.6 estimate |
| Qwen3.5-4B | Q8_0 | 4.5 GB | 50.2 estimate |
| gemma-4-26B-A4B-it | UD-IQ2_M | 10.0 GB | 73.1 estimate |
| MiMo-V2.6-Distill-Qwen-9B | Q8_0 | 9.5 GB | 23.6 estimate |
| gemma-4-E4B-it | Q8_0 | 8.2 GB | 27.5 estimate |
| Qwen3-4B | Q8_0 | 4.3 GB | 52.6 estimate |
| gemma-4-31B-it | UD-IQ2_XXS | 8.5 GB | 26.4 estimate |
| gemma-4-12B-it-qat-q4_0-unquantized | UD-Q4_K_XL | 6.7 GB | 33.5 estimate |
| gemma-3-1b-it | Q8_0 | 1.1 GB | 188 estimate |
| gemma-4-E2B-it | Q8_0 | 5.0 GB | 45.3 estimate |
| DeepSeek-R1-0528-Qwen3-8B | Q8_0 | 8.7 GB | 25.9 estimate |
| Llama-3.2-3B-Instruct | Q8_0 | 3.4 GB | 65.8 estimate |
Similar hardware
NVIDIA RTX 5070 Ti Laptop · NVIDIA RTX A4500 Laptop · NVIDIA RTX 3070 Laptop · NVIDIA RTX 4060 Laptop · NVIDIA RTX 5070 Laptop · NVIDIA RTX 3070 Ti Laptop · NVIDIA RTX 3080 Laptop · NVIDIA RTX 4070 Laptop