A parallel sidekick agent handles localized operations like linting.
A parallel sidekick agent handles localized operations like linting.
@walden_yan You guys are doing an amazing job man!
@eliebakouch BS…
Model routing is a hot topic but there are two challenges to doing routing for coding agents: 1) Even if different models can pass the task, there are subtle differences in behavior & style that mean they aren't perfectly interchangeable. 2) The initial agent prompt isn't enough to know the difficulty of the task. "Fix xyz bug" could be a one-line edge case or could require rearchitecting your entire product; you can't know until you've actually investigated the code. How do you solve these problems? Well, you need evals that account for style and behavior, not just pass/fail. And you need the agent to be able to dynamically update and re-route. We built Devin Fusion with both of these points in mind and found that it reduces costs by 30-40% while still maintaining the frontier intelligence "smell":
Excited to finally share "sidekick" and "mid-session routing" with the world! Our latest methods for multi-model routing. Headlines: - 35% cost savings - Still frontier-level on day-to-day - Works very well with Fable 5 Hope this helps other agent builders out there! https://x.com/walden_yan/status/2071627241818399181/photo/1 https://twitter.com/cognition/status/2071624568465490170
This is a neat idea around doing model routing and sub-agent delegation while ensuring cache hits on accumulated context for all agents. It makes a lot of sense: you want to ensure that all subagents can also make use of accumulated context in their cache https://twitter.com/cognition/status/2071624574270157074
> Devin Fusion builds on this research while avoiding common pitfalls of model routing, like expensive cache misses and lack of generalization. Cognition finally showing serious engineering https://twitter.com/cognition/status/2071624574270157074
Cache misses usually make model routing very expensive. The "sidekick" approach avoids this in several clever ways. https://x.com/walden_yan/status/2071710549633835427/photo/1 https://twitter.com/cognition/status/2071624568465490170
this is how you do a great "model routing" release btw:) https://cognition.com/blog/devin-fusion https://x.com/eliebakouch/status/2071766863328284801/photo/1 https://twitter.com/cognition/status/2071624568465490170
especially with fable-level models, i'm *very* bullish on mid-session routing. (e.g. more routing via sub-agents, etc.) https://twitter.com/walden_yan/status/2071627241818399181
Ask a question below.
Published answers will appear here.
