thinkingmachines/Inkling

Natively multimodal 1T-parameter MoE from Thinking Machines Lab — text/image/audio in, text out, up to 1M context — with relative attention, short convolution, and shared expert sinks.

380 tok/s/user with MTP8 (mean acceptance length 4.5) and 140 tok/s/user without MTP on 4× GB200 GPUs

View on HuggingFaceView on ModelScope