community pool / model answer

Qwen3.5-122B-A10B

chat · Qwen/Qwen3.5-122B-A10B

25.4 tok/s · model median · 10 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
GB10 Grace Blackwell 128GB6-bit17.1 tok/sreported-- GB100,000llama.cpp
GTX 1080 Ti 11GB3-bit3.2 tok/sreported-10.3 GB8,192llama.cpp
Radeon AI Pro R9700 32GB ×34-bit26.5 tok/smeasured1,29677.1 GB65,536llama.cpp
Ryzen AI Max 395 128GB4-bit18 tok/sreported3,41874 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.