M2 Max 96GB vs AMD RX 6700 XT for local LLMs
7 reports on the M2 Max 96GB and 3 on the AMD RX 6700 XT
Which is faster for local LLMs?
On Qwen3 8B at 4-bit, the M2 Max 96GB runs at 43.0 tokens per second (one run) and the AMD RX 6700 XT at 61.0 (one run). 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 7% higher on the M2 Max 96GB (410 GB/s on the M2 Max 96GB, 384 on the AMD RX 6700 XT).
- Memory
- 96GB unified
- Memory bandwidth
- 410 GB/s
- FP16 compute
- 13.6 TFLOPS
- Reports
- 7
- Memory
- 12GB
- Memory bandwidth
- 384 GB/s
- FP16 compute
- 26.4 TFLOPS
- Reports
- 3
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 | M2 Max 96GB | AMD RX 6700 XT |
|---|---|---|
| Qwen3 8B4-bit | 43.0one run | 61.0one run |
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 | M2 Max 96GB | AMD RX 6700 XT |
|---|---|---|
| Qwen3 8B | 17.6 t/sFP16 | 60.3 t/sQ3_K_M |
| Qwen3.8 27B | 5.2 t/sFP16 | 1.6 t/sQ8_0, layers on CPU |
| Qwen3.8 125B · 6B active | 36.2 t/sQ3_K_M | doesn't fit at these settings |
| DeepSeek V4 Flash 284B · 13B active | doesn't fit at these settings | doesn't fit at these settings |
| Qwen3.6 35B · 3B active | 37.6 t/sQ8_0 | 13.3 t/sQ6_K, experts in RAM |
| Qwen3.6 27B | 5.2 t/sFP16 | 1.6 t/sQ8_0, layers on CPU |
| Gemma 4 26B · 4B active | 12.5 t/sFP16 | 62.7 t/sQ2_K |
Change the context, RAM or model in the calculator: M2 Max 96GB · AMD RX 6700 XT
Every report on each card, with its source: M2 Max 96GB · AMD RX 6700 XT