AMD Radeon RX 6900 XT × Llama 3.2 3B Instruct
Llama 3.2 3B Instruct on AMD Radeon RX 6900 XT
55.3best gen tok/s
2,680best prompt tok/s
640max context tested
5results
Measured · all numbers below come from the linked sources
llama.cpp
| Quant | Gen tok/s | Prompt tok/s | Source | Context tested | TTFT | Build |
|---|---|---|---|---|---|---|
| Q4_K - Medium | 55.3 | 2,159 | LocalScore ↗ | 512+128 | 705 ms | llamafile 0.9.2 |
| Q5_K - Medium | 53.6 | 2,339 | LocalScore ↗ | 512+128 | 665 ms | llamafile 0.9.2 |
| Q6_K | 52.8 | 2,124 | LocalScore ↗ | 512+128 | 713 ms | llamafile 0.9.2 |
| Q8_0 | 49.1 | 2,680 | LocalScore ↗ | 512+128 | 603 ms | llamafile 0.9.2 |
| F16 | 29.1 | 1,475 | LocalScore ↗ | 512+128 | 911 ms | llamafile 0.9.2 |
Ollama
No measured Ollama results yet. Submit yours →
MLX
No measured MLX results yet. Submit yours →
"Context tested" is the largest prompt+generation length actually benchmarked (tokens), not a memory-fit estimate.
What else can AMD Radeon RX 6900 XT run? →
Includes results from LocalScore, a Mozilla Builders project by cjpais, mirrored in the cn0303/local-llm-speed-benchmarks dataset. LocalScore runs use llamafile, which is built on llama.cpp.