M5 Ultra 256GB vs NVIDIA RTX 3080 20GB for local LLMs
12 reports on the M5 Ultra 256GB and 9 on the NVIDIA RTX 3080 20GB
Which is faster for local LLMs?
On Qwen3.8 27B at 4-bit, the M5 Ultra 256GB runs at 50.0 tokens per second (one run) and the NVIDIA RTX 3080 20GB at 20.0 (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 1.6x higher on the M5 Ultra 256GB (1,229 GB/s on the M5 Ultra 256GB, 760 on the NVIDIA RTX 3080 20GB).
- Memory
- 256GB unified
- Memory bandwidth
- 1,229 GB/s
- FP16 compute
- not on record
- Reports
- 12
- Memory
- 20GB
- Memory bandwidth
- 760 GB/s
- FP16 compute
- 119 TFLOPS
- Reports
- 9
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 | M5 Ultra 256GB | NVIDIA RTX 3080 20GB |
|---|---|---|
| Qwen3.8 27B4-bit | 50.0one run | 20.0one 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 | M5 Ultra 256GB | NVIDIA RTX 3080 20GB |
|---|---|---|
| Qwen3.8 27B | 14.9 t/sFP16 | 35.4 t/sQ3_K_M |
| Qwen3.8 125B · 6B active | 50.8 t/sQ8_0 | 19.1 t/sQ2_K, experts in RAM |
| DeepSeek V4 Flash 284B · 13B active | 66.9 t/sQ4_K_M | doesn't fit at these settings |
| Qwen3.6 35B · 3B active | 53.5 t/sFP16 | 140.5 t/sQ3_K_S |
| Qwen3.6 27B | 14.9 t/sFP16 | 35.4 t/sQ3_K_M |
| Gemma 4 26B · 4B active | 40.1 t/sFP16 | 79.7 t/sQ4_K_S |
| DeepSeek V4.1 Flash 552B · 16B active | 89.4 t/sQ2_K | doesn't fit at these settings |
Change the context, RAM or model in the calculator: M5 Ultra 256GB · NVIDIA RTX 3080 20GB
Every report on each card, with its source: M5 Ultra 256GB · NVIDIA RTX 3080 20GB