Hardware
CPU only · DDR5-4800
CPU only · DDR5-4800 has 8 / 16 / 32 / 64 / 128 GB System RAM and 76.8 GB/s. See which open models fit, at which quant, and roughly how fast. Figures below use the default memory (32 GB). Updated 2026-10-24.
Memory8 / 16 / 32 / 64 / 128 GB System RAM
Bandwidth76.8 GB/s
Kindcpu
Year2024
Dual-channel DDR5-4800: 2 × 4.8 GT/s × 8 bytes = 76.8 GB/s. Source.
RSS for new models on this device
Models that fit
| Model | Verdict | Quant | Context | tok/s |
|---|---|---|---|---|
| Qwen3.8 35B A3B Distill 35.5B | Runs on CPU | Q5_K_M | 8K | 18–25 |
| Cyber Tiel Coder 35B A3B GGUF MTP 35.5B | Runs on CPU | Q5_K_M (est.) | 8K | 18–25 |
| Qwen3.6 35B A3B 34.7B | Runs on CPU | UD-Q5_K_S | 8K | 18–25 |
| Qwen3.5 35B A3B 34.7B | Runs on CPU | Q5_K_S | 8K | 18–25 |
| Tiel Coder 35B A3B 34.7B | Runs on CPU | UD-Q4_K_XL | 8K | 20–27 |
| Nemotron 3 Nano 30B A3B 31.6B | Runs on CPU | Q5_K_S | 8K | 16–21 |
| NVIDIA Nemotron 3 Nano Omni 30B A3B Reasoning 31.6B | Runs on CPU | UD-Q5_K_S | 8K | 15–21 |
| Qwen3 Coder 30B A3B Instruct 30.5B | Runs on CPU | Q6_K | 8K | 13–18 |
| Qwen3 30B A3B Instruct 2507 30.5B | Runs on CPU | Q6_K | 8K | 13–18 |
| Qwen3 VL 30B A3B Instruct 30.5B | Runs on CPU | Q6_K | 8K | 13–18 |
| Qwen3 30B A3B Thinking 2507 30.5B | Runs on CPU | Q6_K | 8K | 13–18 |
| NVIDIA Nemotron 3.5 Lightning 30B A3B 32.9B | Runs on CPU | UD-Q4_K_XL | 4K | 17–23 |
| LFM2.5 24B A2B 24.0B | Runs on CPU | Q8_0 | 4K | 17–23 |
| Qwen3 Next 80B A3B Instruct 79.7B | Runs on CPU | UD-IQ1_M | 8K | 24–33 |
| Qwen3 Next 80B A3B Thinking 79.7B | Runs on CPU | UD-IQ1_M | 8K | 25–34 |
| GLM 4.7 Flash REAP 23B A3B 23.0B | Runs on CPU | Q8_0 | 8K | 12–16 |
| gpt oss 20b 20.9B | Runs on CPU | UD-Q8_K_XL | 8K | 10–14 |
| Qwen3.6 14B A3B FableVibes 13.8B | Runs on CPU | Q8_0 | 8K | 13–17 |
| LFM2.5 8B A1B 8.5B | Runs on CPU | BF16 | 8K | 18–24 |
| Qwen3.8 2B Distill 1.9B | Runs on CPU | BF16 | 8K | 11–14 |
| Qwen3.5 2B 1.9B | Runs on CPU | BF16 | 8K | 11–15 |
| Hy MT2 1.8B 1.8B | Runs on CPU | Q8_0 | 8K | 17–24 |
| DeepSeek R1 Distill Qwen 1.5B 1.8B | Runs on CPU | BF16 | 8K | 11–15 |
| Qwen3 1.7B 1.7B | Runs on CPU | BF16 | 8K | 9.6–13 |
| Qwen3 VL 2B Instruct 1.7B | Runs on CPU | BF16 | 8K | 9.6–13 |
| Qwen3 VL 2B Thinking 1.7B | Runs on CPU | BF16 | 8K | 9.6–13 |
| Qwen2.5 1.5B Instruct 1.5B | Runs on CPU | F16 | 8K | 13–17 |
| Qwen2.5 Coder 1.5B Instruct 1.5B | Runs on CPU | F16 | 8K | 13–17 |
| Llama 3.2 1B Instruct 1.2B | Runs on CPU | BF16 | 8K | 15–21 |
| LFM2.5 1.2B Instruct 1.2B | Runs on CPU | BF16 | 8K | 16–22 |
| gemma 3 1b it 1.0B | Runs on CPU | BF16 | 8K | 19–26 |
| Qwen3.5 0.8B 0.75B | Runs on CPU | BF16 | 8K | 26–36 |
| Qwen3 0.6B 0.60B | Runs on CPU | BF16 | 8K | 20–27 |
| Qwen2.5 0.5B Instruct 0.49B | Runs on CPU | F16 | 8K | 39–53 |
| Qwen2.5 32B Instruct 32.8B | Runs on CPU | Q5_K_M | 8K | 1.7–2.3 |
| Qwen3 32B 32.8B | Runs on CPU | UD-Q5_K_XL | 8K | 1.7–2.3 |
| Qwen2.5 Coder 32B Instruct 32.8B | Runs on CPU | Q5_K_M | 8K | 1.7–2.3 |
| Qwen3 VL 32B Instruct 32.8B | Runs on CPU | UD-Q5_K_XL | 8K | 1.7–2.3 |
| DeepSeek R1 Distill Qwen 32B 32.8B | Runs on CPU | Q5_K_M | 8K | 1.7–2.3 |
| GLM 4 32B 0414 32.6B | Runs on CPU | UD-Q5_K_XL | 8K | 1.8–2.4 |
105 more models in the catalog do not fit this default profile, or ranked below the first 40 that do.
Example: Runs on CPU on your CPU only · DDR5-4800 at Q5_K_M, 8K context, about 18–25 tok/s. Weights 23.6 GB plus 0.19 GB of KV cache — 24.3 GB of 25.2 GB usable. Up to 16K context still fits. Weights shown in the verdict use GiB (the same 1024³ counting as GPU memory).