community pool / model answer
Llama-3.2-1B-Instruct
chat · meta-llama/Llama-3.2-1B-Instruct
184 tok/s · model median · 7 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| Arc Pro B60 24GB | 8-bit | 189.4 tok/s | reported | 64 | - GB | 4,096 | llama.cpp |
| Arc Pro B60 24GB ×2 | 8-bit | 189.3 tok/s | reported | 64 | - GB | 4,096 | llama.cpp |
| AMD Ryzen 7 7840HS | 2-bit | 91.1 tok/s | reported | - | - GB | 2,048 | llama.cpp |
| AMD Ryzen 7 7840HS | 4-bit | 68.2 tok/s | reported | - | - GB | 2,048 | llama.cpp |
| RTX 5080 16GB | 8-bit | 448.1 tok/s | reported | 14 | - GB | 4,096 | llama.cpp |
Context tested
The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.