community pool / model answer
SmolLM2-135M-Instruct
chat · HuggingFaceTB/SmolLM2-135M-Instruct
407.8 tok/s · model median · 5 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| AMD Ryzen 7 7840HS | 2-bit | 440.2 tok/s | reported | - | - GB | 2,048 | llama.cpp |
| AMD Ryzen 7 7840HS | 4-bit | 375.4 tok/s | reported | - | - GB | 2,048 | llama.cpp |
| AMD Ryzen 7 7840HS | 8-bit | 305.4 tok/s | reported | - | - GB | 2,048 | llama.cpp |
| AMD Ryzen 7 7840HS | 16-bit | 183.5 tok/s | reported | - | - GB | 2,048 | llama.cpp |
| RTX 3060 12GB | 8-bit | 684.5 tok/s | reported | 138 | 1.6 GB | 4,608 | 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.