community pool / model answer

Ling-3.0-flash

chat · inclusionAI/Ling-3.0-flash

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

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Multi-GPU 72GB ×34-bit53.4 tok/smeasured65468.4 GB262,144llama.cpp
Radeon AI Pro R9700 32GB ×34-bit39.3 tok/smeasured-- GB4,096llama.cpp
Ryzen AI Max 395 128GB4-bit39.2 tok/sreported-- GB4,096llama.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.