community pool / model answer

Ternary-Bonsai-27B-gguf

chat · prism-ml/Ternary-Bonsai-27B-gguf

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

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Ryzen AI Max 395 128GB2-bit23.2 tok/smeasured26,689- GB262,144llama.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.