Microsoft · open-weights model
Phi-3.5-mini-instruct
Phi-3.5-mini-instruct is a 3.8B-parameter model by Microsoft. Its smallest quant (Q4_K_M) needs at least 6.1GB of GPU-addressable memory at 8K context; it loads cleanly at a quality quant on 50 of the 50 machines we track.
- Parameters
- 3.8B
- Layers
- 32
- Max context
- 128K
- Attention
- MHA
- License
- mit
Config from microsoft/Phi-3.5-mini-instruct
Quants and what they need
Measured file sizes; “needs” = weights + KV cache at 8K context + overhead.
| Quant | File size | Needs ≥ |
|---|---|---|
| Q8_0 | 4.1 GB | 7.7 GB |
| Q6_K | 3.1 GB | 6.8 GB |
| Q5_K_M | 2.8 GB | 6.5 GB |
| Q4_K_M | 2.4 GB | 6.1 GB |
What runs Phi-3.5-mini-instruct
Every machine we track, graded at its best clean-fit quant.
Verdicts computed from published specs and measured GGUF sizes — how the math works.