Applications
Using the biggest model for every task can work in a demo. In production, cost, latency, and task-specific quality can tell a different stor…
Using the biggest model for every task can work in a demo. In production, cost, latency, and task-specific quality can tell a different story. At ODSC AI West 2026, Rob Ferguson, VP of Technology and
Using the biggest model for every task can work in a demo. In production, cost, latency, and task-specific quality can tell a different story. At ODSC AI West 2026, Rob Ferguson, VP of Technology and Strategy at @FireworksAI_HQ, will present “One Big Model, or Many Small Ones?” Rob will make the case for a different architecture: fine-tuning smaller, task-specific models and coordinating them through routing and evaluation layers. Drawing on experience across Fireworks, Microsoft, and AWS, he’ll explore how teams can identify good candidates for specialized models, build eval sets around their actual workloads, estimate costs more realistically, and route requests to the right model instead of defaulting to one generalist system. Join Rob at ODSC AI West 2026: https://hubs.li/Q04yGpZt0
Related
- Fine-tuning to production inference is the gap where teams get stuck. At #MSBuild today, our own Rob Ferguson, @danielhanchen (@UnslothAI) a…
- Move from test to production by running high-performance inference directly on Foundry. At #MSBuild, we demoed an end-to-end workflow showin…
- Production AI systems place very different demands on infrastructure once workloads scale. How? Join us at #MSBuild to find out. Register he…
- Position: Avoid Overstretching LLMs for every Enterprise Task
- Day 2 at #MSBuild is about what it takes to move beyond generic foundation models. Think customization, inference performance, and getting p…
Source: Fireworks AI (X) | 2026-09-29