M2 Ultra 192GB vs M3 Ultra 96GB for local LLMs
9 reports on the M2 Ultra 192GB and 10 on the M3 Ultra 96GB
Which is faster for local LLMs?
Nobody has reported the same model at the same quant on both cards yet, so there's no measured head-to-head. Memory bandwidth, which caps how fast a card can write, is about the same on both (819 GB/s on the M2 Ultra 192GB, 819 on the M3 Ultra 96GB). The estimates below compare them on the models people run most.
M2 Ultra 192GB
- Memory
- 192GB unified
- Memory bandwidth
- 819 GB/s
- FP16 compute
- 27.2 TFLOPS
- Reports
- 9
M3 Ultra 96GB
- Memory
- 96GB unified
- Memory bandwidth
- 819 GB/s
- FP16 compute
- 21.3 TFLOPS
- Reports
- 10
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.
| Model | M2 Ultra 192GB | M3 Ultra 96GB |
|---|---|---|
| Qwen3.8 27B | 8.3 t/sFP16 | 7.9 t/sFP16 |
| Qwen3.8 125B · 6B active | 33.0 t/sQ6_K | 55.1 t/sQ3_K_M |
| Qwen3.6 35B · 3B active | 29.9 t/sFP16 | 57.1 t/sQ8_0 |
| DeepSeek V4 Flash 284B · 13B active | 62.8 t/sQ3_K_S | doesn't fit at these settings |
| Qwen3.6 27B | 8.3 t/sFP16 | 7.9 t/sFP16 |
| Gemma 4 26B · 4B active | 22.5 t/sFP16 | 19.0 t/sFP16 |
Change the context, RAM or model in the calculator: M2 Ultra 192GB · M3 Ultra 96GB
Every report on each card, with its source: M2 Ultra 192GB · M3 Ultra 96GB