Can Phi 3 mini 128k instruct run on the RTX 5070?
Yes. At 8K context the best fit is Q8_0, using about 7.2 GB of the 11.4 GB usable on the RTX 5070. Estimated speed ceiling: ~165 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) • | 3.3 GB Fits | 4.8 GB Fits | 13.8 GB No | 49.8 GB No | 475 |
| 3-bit (Q3_K_M) • | 3.8 GB Fits | 5.3 GB Fits | 14.3 GB No | 50.3 GB No | 344 |
| 4-bit (Q4_K_M) • | 4.2 GB Fits | 5.7 GB Fits | 14.7 GB No | 50.7 GB No | 281 |
| 5-bit (Q5_K_M) • | 4.6 GB Fits | 6.1 GB Fits | 15.1 GB No | 51.1 GB No | 239 |
| 6-bit (Q6_K) • | 4.9 GB Fits | 6.4 GB Fits | 15.4 GB No | 51.4 GB No | 214 |
| 8-bit (Q8_0) • | 5.7 GB Fits | 7.2 GB Fits | 16.2 GB No | 52.2 GB No | 165 |
| 16-bit (F16/BF16) | 9.1 GB Fits | 10.6 GB Tight | 19.6 GB No | 55.6 GB No | 88 |
• real GGUF file size from bartowski/Phi-3.1-mini-128k-instruct-GGUF. Speed is a bandwidth ceiling (672 GB/s), not a benchmark. Methodology.
More
Phi 3 mini 128k instruct on every GPU and Mac · Every model that runs on the RTX 5070 · Other models that fit 12 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (microsoft/Phi-3-mini-128k-instruct), NVIDIA specs. Data as of 2026-11-17.