NVIDIA P102-100 vs NVIDIA RTX 3060 12GB for local LLMs
2 reports on the NVIDIA P102-100 and 64 on the NVIDIA RTX 3060 12GB
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 22% higher on the NVIDIA P102-100 (440 GB/s on the NVIDIA P102-100, 360 on the NVIDIA RTX 3060 12GB). The estimates below compare them on the models people run most.
NVIDIA P102-100
- Memory
- 10GB
- Memory bandwidth
- 440 GB/s
- FP16 compute
- 10.8 TFLOPS
- Reports
- 2
NVIDIA RTX 3060 12GB
- Memory
- 12GB
- Memory bandwidth
- 360 GB/s
- FP16 compute
- 51 TFLOPS
- Reports
- 64
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 | NVIDIA P102-100 | NVIDIA RTX 3060 12GB |
|---|---|---|
| Qwen3.8 27B | 1.9 t/sQ6_K, layers on CPU | 1.5 t/sQ8_0, layers on CPU |
| Qwen3.8 125B · 6B active | doesn't fit at these settings | doesn't fit at these settings |
| Qwen3.6 35B · 3B active | 16.1 t/sQ5_K_M, experts in RAM | 13.1 t/sQ6_K, experts in RAM |
| DeepSeek V4 Flash 284B · 13B active | doesn't fit at these settings | doesn't fit at these settings |
| Qwen3.6 27B | 1.9 t/sQ6_K, layers on CPU | 1.5 t/sQ8_0, layers on CPU |
| Gemma 4 26B · 4B active | 8.3 t/sQ8_0, experts in RAM | 58.8 t/sQ2_K |
Change the context, RAM or model in the calculator: NVIDIA P102-100 · NVIDIA RTX 3060 12GB
Every report on each card, with its source: NVIDIA P102-100 · NVIDIA RTX 3060 12GB