community pool / model answer

Trinity-Mini

chat · arcee-ai/Trinity-Mini

105 tok/s · model median · 26 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX 3060 12GB ×24-bit99.3 tok/smeasured3,43416.6 GB131,072llama.cpp
RTX 3090 24GB4-bit162.6 tok/smeasured2,136- GB131,072llama.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.