[jumploops]

> For complex tasks, Kimi K2.5 can self-direct an agent swarm with up to 100 sub-agents, executing parallel workflows across up to 1,500 tool calls.

> K2.5 Agent Swarm improves performance on complex tasks through parallel, specialized execution [..] leads to an 80% reduction in end-to-end runtime

Not just RL on tool calling, but RL on agent orchestration, neat!

[storystarling]

1,500 tool calls per task sounds like a nightmare for unit economics though. I've been optimizing my own agent workflows and even a few dozen steps makes it hard to keep margins positive, so I'm not sure how this is viable for anyone not burning VC cash.

[zozbot234]

"tool call" is just a reference to any elementary interaction with the outside system. It's not calling third-party APIs or anything like that.

[storystarling]

True, but that's still 1,500 inference cycles. Even without external API fees, the latency and compute burden seems huge. I don't see how the economics work there without significant subsidies.

[darrinm]

FWIW many tool calls can be and often are made in one inference cycle.

[XCSme]

> Kimi K2.5 can self-direct an agent swarm

Is this within the model? Or within the IDE/service that runs the model?

Because tool calling is mostly just the agent outputting "call tool X", and the IDE does it and returns the data back to AI's context

[mzl]

An LLM model only outputs tokens, so this could be seen as an extension of tool calling where it has trained on the knowledge and use-cases for "tool-calling" itself as a sub-agent.

[XCSme]

Ok, so agent swarm = tool calling where the tool is a LLM call and the argument is the prompt

[IanCal]

Yes largely, although they’ve trained a model specifically for this task rather than using the base model and a bit of prompting.

[dcre]

Sort of. It’s not necessarily a single call. In the general case it would be spinning up a long-running agent with various kinds of configuration — prompts, but also coding environment and which tools are available to it — like subagents in Claude Code.

[mohsen1]

Parallel agents are such a simple, yet powerful hack. Using it in Claude Code with TeammateTool and getting lots of good results!

[esperent]

> TeammateTool

What is this?

[frimmy]

https://x.com/kieranklaassen/status/2014830266515382693 - agent swarms tool shipping w/ cc soon..

[jlu]

claude code hidden feaure currently under a feature flag:

https://github.com/mikekelly/claude-sneakpeek