Tiny but surprisingly capable. Runs on almost anything, even CPU.
chatgeneral · sizes: 1.7B · Apache-2.0
| machine | best fit | verdict | speed |
|---|---|---|---|
| Apple M1 · 8GB | 1.7B FP16 | 🐢 CPU-only (slow) | 1–2 tok/s |
| Apple M1 Pro · 16GB | 1.7B FP16 | ✅ Runs comfortably | 20–35 tok/s |
| Apple M2 · 16GB | 1.7B FP16 | ✅ Runs comfortably | 12–20 tok/s |
| Apple M3 Pro · 18GB | 1.7B FP16 | ✅ Runs comfortably | 17–30 tok/s |
| Apple M2 Pro · 32GB | 1.7B FP16 | ✅ Runs comfortably | 20–35 tok/s |
| Apple M4 Pro · 48GB | 1.7B FP16 | ✅ Runs comfortably | 30–50 tok/s |
| Apple M3 Max · 36GB | 1.7B FP16 | ✅ Runs comfortably | 40–65 tok/s |
| Apple M2 Max · 64GB | 1.7B FP16 | ✅ Runs comfortably | 40–65 tok/s |
| Apple M4 Max · 128GB | 1.7B FP16 | ✅ Runs comfortably | 45–80 tok/s |
| Apple M2 Ultra · 192GB | 1.7B FP16 | ✅ Runs comfortably | 50–85 tok/s |
| NVIDIA GeForce RTX 3060 · 12GB VRAM | 1.7B FP16 | ✅ Runs comfortably | 30–50 tok/s |
| NVIDIA GeForce RTX 4070 · 12GB VRAM | 1.7B FP16 | ✅ Runs comfortably | 45–75 tok/s |
| NVIDIA GeForce RTX 4090 · 24GB VRAM | 1.7B FP16 | ✅ Runs comfortably | 85–150 tok/s |
| NVIDIA GeForce RTX 5080 · 16GB VRAM | 1.7B FP16 | ✅ Runs comfortably | 85–140 tok/s |
| NVIDIA GeForce RTX 5090 · 32GB VRAM | 1.7B FP16 | ✅ Runs comfortably | 150–260 tok/s |
| AMD Radeon RX 7900 XTX · 24GB VRAM | 1.7B FP16 | ✅ Runs comfortably | 80–130 tok/s |
Shopping for a machine? What hardware do I need to run SmolLM2? →
Check your exact rig: open the checker →