Can I Run Llama 3.3 on Framework Desktop (Ryzen AI Max+ 395, 128 GB)?
Superseded model. Llama 3.3 has been superseded by Llama 4. This page is kept for reference; the newer family is a better starting point.
View Llama 4 →
Written by Jakub Rusinowski · Last updated December 8, 2024
Yes — comfortably
Yes, comfortably — Llama 3.3 70B Instruct at Q3_K_M needs about 33.3 GB of the 128 GB usable on Framework Desktop (Ryzen AI Max+ 395, 128 GB), leaving ~94.7 GB spare and running at ~6.1 tok/s (estimated), with room for about 131,072 tokens of context.
Confidence: high · Recommended quantization: Q3_K_M · Estimated speed: ~6.1 tok/s
Framework Desktop (Ryzen AI Max+ 395, 128 GB) — what it gives a model
| Usable memory for models | 128 GB |
| Memory bandwidth | 256 GB/s |
| Form factor | Mini PC |
| Operating system | Windows or Linux |
| Memory upgradeable | No — soldered |
Llama 3.3 on Framework Desktop (Ryzen AI Max+ 395, 128 GB): memory by quantization
| Quant | Memory needed | Fits 128 GB? | Max context | Est. speed | Download |
|---|
| F16 | 143.5 GB | ✗ No | — | — | 140 GB |
| Q8_0 | 77.9 GB | ✓ Yes | 128K | ~2.5 tok/s | 74.4 GB |
| Q6_K | 60.9 GB | ✓ Yes | 128K | ~3.2 tok/s | 57.4 GB |
| Q5_K_M | 53.1 GB | ✓ Yes | 128K | ~3.7 tok/s | 49.6 GB |
| Q4_K_M | 45.7 GB | ✓ Yes | 128K | ~4.4 tok/s | 42.3 GB |
| Q3_K_M | 33.3 GB | ✓ Yes | 128K | ~6.1 tok/s | 29.8 GB |
| Q2_K | 26.5 GB | ✓ Yes | 128K | ~7.7 tok/s | 23 GB |
What to watch out for
- Q3_K_M is the only quantization that fits, and it is a heavily degraded one — expect noticeably worse output than the same model at Q4_K_M. A smaller model at Q4 is usually the better trade.
- Memory on this machine is not upgradeable, so the configuration you buy is the ceiling for every model you will ever run on it.
Framework Desktop 128 GB limitations
- Memory is soldered LPDDR5X — unusually for Framework, this is the one component that cannot be upgraded.
- ROCm/Vulkan support for Strix Halo is younger than CUDA; check your runtime supports it before buying.
Recommended setup
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
How these numbers are calculated
- Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
- 128 GB unified memory at 256 GB/s, shared between CPU and GPU.
- Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
- KV cache is computed from this model's published attention configuration.
FAQ
Can I run Llama 3.3 on Framework Desktop (Ryzen AI Max+ 395, 128 GB)?
Yes, comfortably — Llama 3.3 70B Instruct at Q3_K_M needs about 33.3 GB of the 128 GB usable on Framework Desktop (Ryzen AI Max+ 395, 128 GB), leaving ~94.7 GB spare and running at ~6.1 tok/s (estimated), with room for about 131,072 tokens of context.
Which quantization of Llama 3.3 should I use on Framework Desktop (Ryzen AI Max+ 395, 128 GB)?
Q3_K_M — it needs about 33.3 GB of the 128 GB available, downloads as roughly 29.8 GB, and runs at an estimated 6.1 tokens/sec with up to 128K of context.
What limits Llama 3.3 on Framework Desktop (Ryzen AI Max+ 395, 128 GB)?
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Which runtime should I use?
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Other Computers
Other Models on Framework Desktop (Ryzen AI Max+ 395, 128 GB)
Llama 3.3 on GPUs
What This Model Is Good At
Model & Tools
← Can I Run It? | Llama 3.3 model page | Check your hardware