phi-3-medium 14B
phi-3-medium 14B is an open-weights language model from Microsoft, released April 23, 2024. It has 14B parameters (dense). On the Epoch Capabilities Index it scores 114.3, rank 193 of 256 rated models.
Check it against your own PC →
Can you run phi-3-medium 14B locally?
The smallest common GPU it runs on is 6 GB of VRAM (with 32 GB of RAM): Q3_K_M, split GPU / CPU, roughly 6 tokens per second. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Runs, slowly | Q3_K_M (okay) | 5.2 GB | 4.8 GB | ~6 |
| 8 GB VRAM | Runs, slowly | Q4_K_M (very good) | 7.2 GB | 4.5 GB | ~6 |
| 12 GB VRAM | Runs on GPU | IQ4_XS (good) | 10.6 GB | – | ~28 |
| 16 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 14.7 GB | – | ~25 |
| 24 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 14.7 GB | – | ~47 |
| 32 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 14.7 GB | – | ~86 |
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Knowledge | 27.9 | GPQA diamond, MMLU |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| MMLU | 70.7% |
| MATH level 5 | 17.6% |
| GPQA Diamond | 3.6% |
Links
Models with a similar score
Claude 3 Haiku · Claude 2 · phi-3-small 7.4B · Claude Instant · PaLM 2-L · Qwen2.5-Coder-14B · Mistral NeMo · Mixtral 8x7B
More from Microsoft
MAI-Cyber-1.1-Flash · MAI-Base-2 · MAI-Cyber-1-Flash · MAI-Code-1.1-Flash · Mage VL · MAI-Thinking-1 · MAI-Code-1-Flash · Phi-4