[bob1029]

This feels like the dialup->broadband transition to me.

I was already a huge proponent of Luna for things like deep research. Being able to run 5x more for the same cost is simply bananas. We are already running 10 parallel agents for hypothesis generation. I cannot imagine 50. The statistics become much more interesting & powerful when you can run so many samples of the exact same prompt+model without breaking the bank.

[andai]

Do you have a sense of which tasks benefit from more agents and which don't?

[bob1029]

Anything related to reading and interpreting the environment seems to always benefit from the addition of more agents to the search party, assuming you have some rational way to synthesize their results.

Taking actions that mutate the environment is a different story. I think this is where you run into diminishing returns very quickly. You generally want one strong agent to act given the results of all the searching that was done. If the plan is clear, you don't need a genius model to execute it.

[handfuloflight]

I definitely think you want the genius model to synthesize everything that rolls up to them.

[andai]

I think this is an unsolved problem. The most interesting thing I saw here is the Recursive Language Models paper.

https://arxiv.org/abs/2512.24601

There's also a great write up here by the author:

https://alexzhang13.github.io/blog/2025/rlm/

[bob1029]

I used this architecture for a while. The problem I have with it is that starting from one agent and fanning out keeps things mostly aligned with that single reasoning trajectory, even as you get a few layers into the stack. Every recursive invocation is a product of the caller's current state. Diversity doesn't really occur on its own unless the environment (tool calling) is complex/chaotic.

RLM might be more useful on the execution side than on the research side. In fact, these somehow feel like they might be exact inverses of each other in terms of what the ideal architecture looks like. At some point you definitely do need something in the middle that has it all sorted out.

[andai]

Yeah, a lot of people have reported significant gains from "model fusion" lately, which I think the literature just calls ensembles. You just ask a bunch of models to do a thing and then you combine the results.

Research tasks benefit the most from this. Because such work benefits from having a large number of agents working on a problem in parallel (i.e. crawling the web), and the model size becomes less important past a certain minimum.

I don't know about other categories of work, like programming. I imagine looking for bugs or security issues would benefit from it.

[kridsdale1]

This is why in your brain you have trillion threads processing and summarizing sensor data (immutable functions), but a SINGLE thread of “execution” which we call the conscious soul.

[Imanari]

How do you run 'deep research'?

[dannyw]

Deep research is basically a LLM with web search, and a "work really hard" goal-orientated prompt, and some output formatting suggestions.

[kridsdale1]

And self forking fan out.

[wahnfrieden]

They’re using the term colloquially. Deep research afaik is an outdated specific model.

[cg5280]

It's a feature offered in ChatGPT and other platforms, though probably gated behind paid subscriptions.

[fy20]

Used to be part of the $20/mo plan but it's not anymore (not sure if they removed it completely). However GPT-5.6 is pretty good at researching if you prompt it right, I've regularly had it spend 5+ minutes researching topic with lots of web searches.

[kaufmann]

They placed it in the Plugins submenu.

[jrflo]

Very interesting. Can you share more about your hypothesis/research pipeline? I have been using Sol for those types of task because I figured you'd need more reasoning for getting good ideas, but maybe quantity > quality at a certain point?

[bob1029]

Here is a rough approximation of the pipeline I use:

Phase 1 - Run X copies of Luna in parallel over the user's prompt. The purpose is to generate a diverse set of hypotheses.

Phase 2 - Run Y copies of Terra in parallel to investigate the hypothesis results, with each receiving them in a randomized order.

Phase 3 - Run 1 copy of Sol over investigation reports.

The goal is to ensure that the agent covers more initial starting points before presenting a final conclusion. If you only run a single copy of Sol and it hooks onto something wrong, it might not recover.

[eevmanu]

How do you run this phases and parallelization on each?

Via just ... "prompting it"?

Or do you use any tool in the middle to ensure this agent architecture?

Just curious if there is any workflow-like tool in the middle that is helping.

[bob1029]

There is a deterministic framework that enforces the process. I use the raw chat completion APIs. No "reasoning effort" is applied on top of these requests (it's not even allowed anymore), so there's actually no way to influence this via prompting.

[AaronAPU]

Does it actually yield results better than just sending it to 5.6 Pro?