Written by Jakub Rusinowski · Last updated March 12, 2025
Yes — comfortably
Yes, comfortably — Gemma 3 27B Instruct at Q8_0 needs about 37.8 GB of the 128 GB usable on Framework Desktop (Ryzen AI Max+ 395, 128 GB), leaving ~90.2 GB spare and running at ~5.8 tok/s (estimated), with room for about 65,536 tokens of context.
Confidence: high · Recommended quantization: Q8_0 · Estimated speed: ~5.8 tok/s
| Usable memory for models | 128 GB |
| Memory bandwidth | 256 GB/s |
| Form factor | Mini PC |
| Operating system | Windows or Linux |
| Memory upgradeable | No — soldered |
| Quant | Memory needed | Fits 128 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 63.1 GB | ✓ Yes | 64K | ~3.3 tok/s | 54 GB |
| Q8_0 | 37.8 GB | ✓ Yes | 64K | ~5.8 tok/s | 28.7 GB |
| Q6_K | 31.3 GB | ✓ Yes | 64K | ~7.2 tok/s | 22.1 GB |
| Q5_K_M | 28.3 GB | ✓ Yes | 64K | ~8.1 tok/s | 19.1 GB |
| Q4_K_M | 25.4 GB | ✓ Yes | 64K | ~9.2 tok/s | 16.3 GB |
| Q3_K_M | 20.6 GB | ✓ Yes | 64K | ~11.9 tok/s | 11.5 GB |
| Q2_K | 18 GB | ✓ Yes | 64K | ~14.3 tok/s | 8.9 GB |
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| Gemma 3 27B Instruct | 25.4 GB | ✓ Fits | ~9.2 tok/s |
| Gemma 3 12B Instruct | 11.1 GB | ✓ Fits | ~20.8 tok/s |
| Gemma 3 4B Instruct | 4.4 GB | ✓ Fits | ~56.2 tok/s |
| Gemma 3 1B Instruct | 2 GB | ✓ Fits | ~143.6 tok/s |
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Yes, comfortably — Gemma 3 27B Instruct at Q8_0 needs about 37.8 GB of the 128 GB usable on Framework Desktop (Ryzen AI Max+ 395, 128 GB), leaving ~90.2 GB spare and running at ~5.8 tok/s (estimated), with room for about 65,536 tokens of context.
Q8_0 — it needs about 37.8 GB of the 128 GB available, downloads as roughly 28.7 GB, and runs at an estimated 5.8 tokens/sec with up to 64K of context.
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving