AI inference | Tech
Tech news, context and sources.
Latest articles
TokenGate uses learned token selection to accelerate long-context prefill
Reported time-to-first-token gains require model adaptation and task-specific accuracy checks, rather than a universal plug-in speedup.
Cerebras explains multi-wafer inference for trillion-parameter models
Weights stay on-chip while activations move between wafers; the 59-wafer mapping is an illustration with specific quantization assumptions.
You’ve reached the end.