community pool / model answer

SmolLM2-360M-Instruct

chat · HuggingFaceTB/SmolLM2-360M-Instruct

189.9 tok/s · model median · 4 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
AMD Ryzen 7 7840HS2-bit208 tok/sreported-- GB2,048llama.cpp
AMD Ryzen 7 7840HS4-bit171.8 tok/sreported-- GB2,048llama.cpp
AMD Ryzen 7 7840HS8-bit130.6 tok/sreported-- GB2,048llama.cpp
RTX 3060 12GB8-bit437 tok/sreported2291.9 GB4,608llama.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.