M5 Ultra 96GB vs NVIDIA RTX 3090 for local LLMs
3 reports on the M5 Ultra 96GB and 165 on the NVIDIA RTX 3090
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 31% higher on the M5 Ultra 96GB (1,229 GB/s on the M5 Ultra 96GB, 936 on the NVIDIA RTX 3090). The estimates below compare them on the models people run most.
M5 Ultra 96GB
- Memory
- 96GB unified
- Memory bandwidth
- 1,229 GB/s
- FP16 compute
- not on record
- Reports
- 3
NVIDIA RTX 3090
- Memory
- 24GB
- Memory bandwidth
- 936 GB/s
- FP16 compute
- 142 TFLOPS
- Reports
- 165
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 | M5 Ultra 96GB | NVIDIA RTX 3090 |
|---|---|---|
| Qwen3.8 27B | 14.9 t/sFP16 | 33.9 t/sQ4_K_M |
| Qwen3.8 125B · 6B active | 103.1 t/sQ3_K_M | 19.4 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 | 107.0 t/sQ8_0 | 120.1 t/sQ4_K_S |
| Qwen3.6 27B | 14.9 t/sFP16 | 33.9 t/sQ4_K_M |
| Gemma 4 26B · 4B active | 35.5 t/sFP16 | 76.4 t/sQ5_K_S |
Change the context, RAM or model in the calculator: M5 Ultra 96GB · NVIDIA RTX 3090
Every report on each card, with its source: M5 Ultra 96GB · NVIDIA RTX 3090