Apple M2 Ultra (76-core GPU, 192GB) × Gemma 3 4B
Can the Apple M2 Ultra (76-core GPU, 192GB) run Gemma 3 4B?
Yes
Yes. The Apple M2 Ultra (76-core GPU, 192GB) can run Gemma 3 4B at up to F16 (needs ~10.8 GB of 144 GB usable), at an estimated ~138 tokens/sec at Q4_K_M.
Speed is estimated from memory bandwidth calibrated with a measured 94.27 tok/s Llama 2 7B Q4_0 result (source). No direct public benchmark of this exact pair is recorded yet.
By quantization (8K context)
| Quant | Memory needed | Fits? | Est. tok/s |
|---|---|---|---|
| Q3_K_M | 4.1 GB | ✅ | ~172 |
| Q4_K_M | 4.6 GB | ✅ | ~138 |
| Q5_K_M | 5.1 GB | ✅ | ~118 |
| Q6_K | 5.6 GB | ✅ | ~102 |
| Q8_0 | 6.6 GB | ✅ | ~78.9 |
| F16 | 10.8 GB | ✅ | ~41.9 |
Fastest hardware for Gemma 3 4B
- NVIDIA GeForce RTX 5090 — ~300 tok/s at Q4_K_M
- NVIDIA GeForce RTX 4090 — ~273 tok/s at Q4_K_M
- AMD Radeon RX 7900 XTX — ~270 tok/s at Q4_K_M
- NVIDIA GeForce RTX 5080 — ~267 tok/s at Q4_K_M
- NVIDIA GeForce RTX 5070 Ti — ~259 tok/s at Q4_K_M
No suitable GPU? Rent one by the hour on Vast.ai or RunPod (affiliate links).
Get new models & benchmarks in your inbox
No spam. Unsubscribe anytime by replying to any email.