Meta's workhorse general model. Reliable all-rounder with 128K context.
chatgeneralreasoningagents · sizes: 8B, 70B · Llama 3.1 Community
| machine | best fit | verdict | speed |
|---|---|---|---|
| Apple M1 · 8GB | 8B Q5_K_M | 🐢 CPU-only (slow) | 0.9–1 tok/s |
| Apple M1 Pro · 16GB | 8B Q6_K | ✅ Runs (tight) | 16–25 tok/s |
| Apple M2 · 16GB | 8B Q6_K | ✅ Runs (tight) | 9–15 tok/s |
| Apple M3 Pro · 18GB | 8B Q6_K | ✅ Runs (tight) | 13–20 tok/s |
| Apple M2 Pro · 32GB | 8B Q8_0 | ✅ Runs comfortably | 13–20 tok/s |
| Apple M4 Pro · 48GB | 70B Q2_K | ✅ Runs (tight) | 5–9 tok/s |
| Apple M3 Max · 36GB | 8B Q8_0 | ✅ Runs comfortably | 20–35 tok/s |
| Apple M2 Max · 64GB | 70B Q4_K_M | ✅ Runs (tight) | 4–7 tok/s |
| Apple M4 Max · 128GB | 70B Q8_0 | ✅ Runs (tight) | 3–5 tok/s |
| Apple M2 Ultra · 192GB | 70B Q8_0 | ✅ Runs comfortably | 4–7 tok/s |
| NVIDIA GeForce RTX 3060 · 12GB VRAM | 8B Q8_0 | ✅ Runs (tight) | 17–30 tok/s |
| NVIDIA GeForce RTX 4070 · 12GB VRAM | 8B Q8_0 | ✅ Runs (tight) | 25–40 tok/s |
| NVIDIA GeForce RTX 4090 · 24GB VRAM | 8B Q8_0 | ✅ Runs comfortably | 50–85 tok/s |
| NVIDIA GeForce RTX 5080 · 16GB VRAM | 8B Q8_0 | ✅ Runs comfortably | 50–80 tok/s |
| NVIDIA GeForce RTX 5090 · 32GB VRAM | 70B Q2_K | ✅ Runs (tight) | 30–45 tok/s |
| AMD Radeon RX 7900 XTX · 24GB VRAM | 8B Q8_0 | ✅ Runs comfortably | 45–75 tok/s |
Shopping for a machine? What hardware do I need to run Llama 3.1? →
Check your exact rig: open the checker →
20 models share Llama 3.1's KV-cache geometry — they size memory the same way, so a rig that fits one tends to fit its same-size kin.
…and 8 more.
Explore the whole catalog by architecture: open the graph →