AMD Radeon Pro V620 32GB vs AMD Radeon Pro W7800 32GB for local LLMs
12 reports on the AMD Radeon Pro V620 32GB and 0 on the AMD Radeon Pro W7800 32GB
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 13% higher on the AMD Radeon Pro W7800 32GB (512 GB/s on the AMD Radeon Pro V620 32GB, 576 on the AMD Radeon Pro W7800 32GB). The estimates below compare them on the models people run most.
- Memory
- 32GB
- Memory bandwidth
- 512 GB/s
- FP16 compute
- 41.5 TFLOPS
- Reports
- 12
- Memory
- 32GB
- Memory bandwidth
- 576 GB/s
- FP16 compute
- 90.4 TFLOPS
- Reports
- 0
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 Radeon Pro V620 32GB | AMD Radeon Pro W7800 32GB |
|---|---|---|
| Qwen3.8 27B | 12.8 t/sQ6_K | 14.4 t/sQ6_K |
| Qwen3.8 125B · 6B active | 18.1 t/sQ2_K, experts in RAM | 18.4 t/sQ2_K, experts in RAM |
| Qwen3.6 35B · 3B active | 46.6 t/sQ5_K_M | 52.4 t/sQ5_K_M |
| DeepSeek V4 Flash 284B · 13B active | doesn't fit at these settings | doesn't fit at these settings |
| Qwen3.6 27B | 12.8 t/sQ6_K | 14.4 t/sQ6_K |
| Gemma 4 26B · 4B active | 32.7 t/sQ6_K | 36.8 t/sQ6_K |
Change the context, RAM or model in the calculator: AMD Radeon Pro V620 32GB · AMD Radeon Pro W7800 32GB
Every report on each card, with its source: AMD Radeon Pro V620 32GB · AMD Radeon Pro W7800 32GB