AMD Strix Halo 128GB vs NVIDIA Tesla P100 16GB for local LLMs
82 reports on the AMD Strix Halo 128GB and 8 on the NVIDIA Tesla P100 16GB
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 2.9x higher on the NVIDIA Tesla P100 16GB (256 GB/s on the AMD Strix Halo 128GB, 732 on the NVIDIA Tesla P100 16GB). The estimates below compare them on the models people run most.
- Memory
- 128GB unified
- Memory bandwidth
- 256 GB/s
- FP16 compute
- 59 TFLOPS
- Reports
- 82
- Memory
- 16GB
- Memory bandwidth
- 732 GB/s
- FP16 compute
- 19 TFLOPS
- Reports
- 8
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 | AMD Strix Halo 128GB | NVIDIA Tesla P100 16GB |
|---|---|---|
| Qwen3.8 27B | 3.5 t/sFP16 | 47.7 t/sQ2_K |
| Qwen3.8 125B · 6B active | 12.6 t/sQ6_K | doesn't fit at these settings |
| DeepSeek V4 Flash 284B · 13B active | 33.5 t/sQ2_K | doesn't fit at these settings |
| Qwen3.6 35B · 3B active | 12.6 t/sFP16 | 157.9 t/sQ2_K |
| Qwen3.6 27B | 3.5 t/sFP16 | 47.7 t/sQ2_K |
| Gemma 4 26B · 4B active | 9.5 t/sFP16 | 106.4 t/sQ3_K_S |
Change the context, RAM or model in the calculator: AMD Strix Halo 128GB · NVIDIA Tesla P100 16GB
Every report on each card, with its source: AMD Strix Halo 128GB · NVIDIA Tesla P100 16GB