According to Liquid AI's CEO, the primary application of architectural research has become enabling efficiency—reducing cost, latency, and memory without sacrificing quality. The next major breakthroughs in AI *capability* are more likely to stem from new learning algorithms and data paradigms rather than architecture alone.

As AI models scale, their optimal architecture changes. Smaller models benefit from architectural "biases" like gating for efficiency. However, at massive scale (trillions of parameters), unstructured architectures like Transformers, which rely on simple matrix multiplication, become superior because they scale with fewer constraints.

Liquid AI uses an automated system to discover neural architectures, avoiding human bias. Crucially, it bypasses misleading proxy metrics like perplexity by putting the target hardware in the loop and evaluating models directly on the customer's downstream tasks, optimizing for latency, memory, and quality.

Successful AI models will be small, specialized ones that run efficiently on consumer CPUs at the edge (laptops, phones). This leverages existing hardware (e.g., Apple's M-series chips) and avoids costly cloud GPUs, creating a strategic advantage for companies like Apple.

Model architecture decisions directly impact inference performance. AI company Zyphra pre-selects target hardware and then chooses model parameters—such as a hidden dimension with many powers of two—to align with how GPUs split up workloads, maximizing efficiency from day one.

Chinese AI models like Kimi achieve dramatic cost reductions through specific architectural choices, not just scale. Using a "mixture of experts" design, they only utilize a fraction of their total parameters for any given task, making them far more efficient to run than the "dense" models common in the West.

To operate efficiently under power and compute constraints, edge AI systems use a pipeline approach. A simple, low-power model runs continuously for initial detection, only activating a more complex, power-intensive model when a specific event or object of interest is identified.

Input-dependent gating, where a model's transformation is modified based on the current input, is a powerful theme in architectures like Mamba and Liquid AI's LFM. This allows the model to learn adaptable *dynamics* during backpropagation, not just static parameters, which is key for efficiency and generalization.

A cost-effective AI architecture involves using a small, local model on the user's device to pre-process requests. This local AI can condense large inputs into an efficient, smaller prompt before sending it to the expensive, powerful cloud model, optimizing resource usage.