M1 Max 32GB vs NVIDIA RTX 4090 for local LLMs
2 reports on the M1 Max 32GB and 36 on the NVIDIA RTX 4090
Which is faster for local LLMs?
On Qwen3.8 27B at 4-bit, the M1 Max 32GB runs at 15.8 tokens per second (one run) and the NVIDIA RTX 4090 at 31.5 (2 runs). Those are community reports whose engines and settings differ, so the gap is a rough sign, not a controlled measure of the two cards. Memory bandwidth, which caps how fast a card can write, is 2.5x higher on the NVIDIA RTX 4090 (410 GB/s on the M1 Max 32GB, 1,008 on the NVIDIA RTX 4090).
- Memory
- 32GB unified
- Memory bandwidth
- 410 GB/s
- FP16 compute
- 10.4 TFLOPS
- Reports
- 2
- Memory
- 24GB
- Memory bandwidth
- 1,008 GB/s
- FP16 compute
- 330 TFLOPS
- Reports
- 36
Measured on both
Median tokens per second from plain runs (one device, one request, no speculative decoding, the whole model in the device's memory) of the same model size at the same quant level. Engines and context lengths can still differ between the runs. How to read these
| Model | M1 Max 32GB | NVIDIA RTX 4090 |
|---|---|---|
| Qwen3.8 27B4-bit | 15.8one run | 31.52 runs |
Estimated by the calculator
The quant the calculator recommends for each card and its estimated speed at a 32,768-token context with 32 GB of system RAM, from memory bandwidth and the model's shape. Where a model was also measured above, the measurement wins.
| Model | M1 Max 32GB | NVIDIA RTX 4090 |
|---|---|---|
| Qwen3.8 27B | 20.1 t/sQ4_K_S | 36.5 t/sQ4_K_M |
| Qwen3.8 125B · 6B active | doesn't fit at these settings | 19.6 t/sQ2_K, experts in RAM |
| DeepSeek V4 Flash 284B · 13B active | doesn't fit at these settings | doesn't fit at these settings |
| Qwen3.6 35B · 3B active | 73.2 t/sQ3_K_M | 129.3 t/sQ4_K_S |
| Qwen3.6 27B | 20.1 t/sQ4_K_S | 36.5 t/sQ4_K_M |
| Gemma 4 26B · 4B active | 44.3 t/sQ4_K_M | 82.3 t/sQ5_K_S |
| DeepSeek V4.1 Flash 552B · 16B active | doesn't fit at these settings | doesn't fit at these settings |
Change the context, RAM or model in the calculator: M1 Max 32GB · NVIDIA RTX 4090
Every report on each card, with its source: M1 Max 32GB · NVIDIA RTX 4090