community pool / model answer

Nex-N2-mini

chat · nex-agi/Nex-N2-mini

104.5 tok/s · model median · 9 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Arc Pro B70 32GB4-bit104.5 tok/smeasured-26.5 GB2,048llama.cpp
RTX 3050 8GB4-bit40.5 tok/sreported74- GB28,000llama.cpp
RX 6600 4GB ×24-bit40.5 tok/sreported74- GB28,000llama.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.