AMD Radeon Pro W6800 vs AMD RX 7900 XTX for local LLMs
0 reports on the AMD Radeon Pro W6800 and 28 on the AMD RX 7900 XTX
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 1.9x higher on the AMD RX 7900 XTX (512 GB/s on the AMD Radeon Pro W6800, 960 on the AMD RX 7900 XTX). The estimates below compare them on the models people run most.
AMD Radeon Pro W6800
- Memory
- 32GB
- Memory bandwidth
- 512 GB/s
- FP16 compute
- 35.7 TFLOPS
- Reports
- 0
AMD RX 7900 XTX
- Memory
- 24GB
- Memory bandwidth
- 960 GB/s
- FP16 compute
- 123 TFLOPS
- Reports
- 28
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 W6800 | AMD RX 7900 XTX |
|---|---|---|
| Qwen3.8 27B | 12.8 t/sQ6_K | 34.8 t/sQ4_K_M |
| Qwen3.8 125B · 6B active | 18.1 t/sQ2_K, experts in RAM | 19.5 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 | 46.6 t/sQ5_K_M | 123.2 t/sQ4_K_S |
| Qwen3.6 27B | 12.8 t/sQ6_K | 34.8 t/sQ4_K_M |
| Gemma 4 26B · 4B active | 32.7 t/sQ6_K | 78.4 t/sQ5_K_S |
Change the context, RAM or model in the calculator: AMD Radeon Pro W6800 · AMD RX 7900 XTX
Every report on each card, with its source: AMD Radeon Pro W6800 · AMD RX 7900 XTX