GPUs / AMD
RX 6900 XT
- Memory
- GDDR6
- Bus
- 256-bit
- Published peak
- 512 GB/s
- Capacity
- 16 GB
Measured
- Read
- 502.8GB/s
- Write
- 464.0GB/s
- Copy
- 426.9GB/s
- Random
- 76.4GB/s
- Latency
- 245ns
- Sustain
- 426.9GB/s
Medians of 1 verified run from 1 operator. Read reaches 98 % of the published peak.
Local AI speed
| Model | Parameters | 4-bit, tokens/s | 8-bit, tokens/s |
|---|---|---|---|
| Llama 3.1 8B | 8.03B | ≤ 125 | ≤ 63 |
| Qwen 2.5 14B | 14.7B | ≤ 68 | ≤ 34 |
| gpt-oss-20b | 21B, 3.6B active | ≤ 279 | Does not fit |
| Mistral Small 24B | 24B | ≤ 42 | Does not fit |
| Qwen 2.5 32B | 32.5B | Does not fit | Does not fit |
| Llama 3.3 70B | 70.6B | Does not fit | Does not fit |
| gpt-oss-120b | 117B, 5.1B active | Does not fit | Does not fit |