🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN ON APPLE
Run Llama 2-70B on Apple hardware
13 options ranked by estimated speed at Q4 / 8k context.
| Hardware | Est tok/s | Fits 8k? | Max ctx | ~Price |
|---|---|---|---|---|
| Apple M2 Ultra 192GB → | ~16.7 | yes | 403k | ~$6,600 |
| Apple M2 Ultra 64GB → | ~16.7 | yes | 52k | ~$4,000 |
| Apple M1 Ultra 128GB → | ~16.7 | yes | 227k | ~$4,800 |
| Apple M1 Ultra 64GB → | ~16.7 | yes | 52k | ~$3,700 |
| Apple M4 Max 64GB → | ~11.4 | yes | 52k | ~$3,900 |
| Apple M4 Max 128GB → | ~11.4 | yes | 227k | ~$4,900 |
| Apple M4 Max 48GB → | ~11.4 | tight | 8k | ~$3,300 |
| Apple M3 Max 48GB → | ~8.4 | tight | 8k | ~$3,200 |
| Apple M3 Max 64GB → | ~8.4 | yes | 52k | ~$3,700 |
| Apple M3 Max 128GB → | ~8.4 | yes | 227k | ~$4,700 |
| Apple M1 Max 64GB → | ~8.4 | yes | 52k | ~$2,600 |
| Apple M2 Max 64GB → | ~8.4 | yes | 52k | ~$2,800 |
| Apple M2 Max 96GB → | ~8.4 | yes | 139k | ~$3,300 |
Estimate at Q4 (memory-bound). Click any option for its full per-hardware estimate and measured numbers.
FAQ
What Apple hardware runs Llama 2-70B?▶
On GenAIList, 13 Apple options can run Llama 2-70B at Q4 — the fastest is the Apple M2 Ultra 192GB at ~16.7 tok/s.
How fast is Llama 2-70B on Apple hardware?▶
Estimated generation speed ranges from ~8.4 to ~16.7 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.