Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Open Deep Research

github.com

Post reply on HN
Print view View original

31–40 of 83 posts

Re: Open Deep Research

#31

Post by aubanel »

Hi all! Aymeric (m-ric) here, maintainer of smolagents and part of the team who built this. Happy to see this interesting people here!

Few points:

- open Deep Research is not a production app, but it could easily be productionized (would need to be faster + good UX).

- As the GAIA score of 55% (not 54%, that would be lame) says, it's not far from the Deep Research score of 67%. It's also not there yet: I think the main point of progress is to improve web browsing. We're working on integrating vision models (for now we've used a text browser developed by the Microsofit autogen team, congrats to them) because it's probably the best way to really interact with webpages.

- Open Deep Research is built on smolagents, a library that we're building, for which the core is having agents that write their actions (tool calls) in code snippets instead of the unpractical JSON blobs + parsing that everyone incl OpenAI and Anthropic use for their agentic/tool-calling APIs. Don't hesitate to go try out the lib and drop issues/PRs!

- smolagents does code execution, which means "danger for your machine" if ran locally. We've railguardeed that a bit with our custom python interpreter, but it will never be 100% safe, so we're enabling remote execution with E2B and soon Docker.

Re: Open Deep Research

#32

Post by lhl »

transpute wrote: post #20
rvz wrote: post #13

Of course. The first of many open source versions of 'Deep Research' projects are now appearing as predicted [0] but in less than a month. Faster than expected. Open source is already at the finish line. [0] https://news.ycombinator.com/item?id=42913379

> Nothing that Perplexity + DeepSeek-R1 can already do Any public comparisons of OAI Deep Research report quality with Perplexity + DeepSeek-R1, on the same query? How do cost and query limits compare?

I've been using GenSpark.ai for the past month to do research (its agents usually does ~20 minutes, but I've seen it go up to almost 2 hours on a task) - it uses a Mixture of Agents approach using GPT-4o, Claude 3.5 Sonnet and Gemini 1.5 Pro and searches for hundreds of sources.

I reran some of these searches and I've so far found OpenAI Deep Research to be superior for technical tasks. Here's one example:

https://chatgpt.com/share/67a10f6d-28cc-8012-bf98-05dcdb705c... vs https://www.genspark.ai/agents?id=c896d5bc-321b-46ca-9aaa-62...

I've been giving Deep Research a good workout, although I'm still mystified if switching between the different base model matters, besides o1 pro always seeming to fail to execute the Deep Research tool.

Re: Open Deep Research

#33

Post by davidsojevic »

aubanel wrote: post #31

Hi all! Aymeric (m-ric) here, maintainer of smolagents and part of the team who built this. Happy to see this interesting people here! Few points: - open Deep Research is not a production app, but it could easily be productionized (would need to be faster + good UX). - As the GAIA score of 55% (not 54%, that would be lame) says, it's not far from the Deep Research score of 67%. It's also not there yet: I think the ma…

Great work on this, Aymeric and team! In terms of improving browsing and/or data sources, do you think it might be worth integrating things like Google Scholar search capability to increase the depth of some of the research that can be done?

It's something I'd be happy to explore a bit if it's of interest.

Re: Open Deep Research

#35

Post by aubanel »

davidsojevic wrote: post #33
aubanel wrote: post #31

Hi all! Aymeric (m-ric) here, maintainer of smolagents and part of the team who built this. Happy to see this interesting people here! Few points: - open Deep Research is not a production app, but it could easily be productionized (would need to be faster + good UX). - As the GAIA score of 55% (not 54%, that would be lame) says, it's not far from the Deep Research score of 67%. It's also not there yet: I think the ma…

Great work on this, Aymeric and team! In terms of improving browsing and/or data sources, do you think it might be worth integrating things like Google Scholar search capability to increase the depth of some of the research that can be done? It's something I'd be happy to explore a bit if it's of interest.

That's a good idea! Could be a very nice tool to add to the lib!

Re: Open Deep Research

#36

Post by transpute »

aubanel wrote: post #31

Hi all! Aymeric (m-ric) here, maintainer of smolagents and part of the team who built this. Happy to see this interesting people here! Few points: - open Deep Research is not a production app, but it could easily be productionized (would need to be faster + good UX). - As the GAIA score of 55% (not 54%, that would be lame) says, it's not far from the Deep Research score of 67%. It's also not there yet: I think the ma…

> smolagents does code execution, which means "danger for your machine" if ran locally. We've railguardeed that a bit with our custom python interpreter, but it will never be 100% safe, so we're enabling remote execution with E2B and soon Docker.

Those remote interfaces may also work with local VMs for isolation.

Re: Open Deep Research

#37

Post by energy123 »

bossyTeacher wrote: post #22

So basically, Altman announced Deep Research less than a month ago and open-source alternatives are already out? Investors are not going to be happy unless OpenAI outperforms them all by an order of magnitude

You're being slippery with language. The open source version doesn't get 26% on Humanity's Last Exam. It's not "already out".

Re: Open Deep Research

#38

Post by anon373839 »

tptacek wrote: post #17
cma wrote: post #12

Earlier quoted context omitted.

The rise of captchas on regular content, no longer just for posting content, could ruin this. Cloudflare and other companies have set things up to go through a few hand selected scrapers and only they will be able to offer AI browsing and research services.

I think the opposite problem is going to occur with captchas for whatever it's worth: LLMs are going to obsolete them. It's an arms race where the defender has a huge constraint the attacker doesn't (pissing off real users); in that way, it's kind of like the opposite dynamics that password hashes exploit.

I’m not sure about that. There’s a lot of runway left for obstacles that are easy for humans and hard/impossible for AI, such as direct manipulation puzzles. (AI models have latency that would be impossible to mask.) On the other hand, a11y needs do limit what can be lawfully deployed…

Re: Open Deep Research

#39

Post by gmerc »

flaviuspopan wrote: post #28

I signed up for Gemini Advanced to get access to Deep Research on Feb 1st and felt on top of the world. So cool, so advanced. Then OpenAI announced theirs on the 2nd: https://openai.com/index/introducing-deep-research/ Ethan Mollick called Google's undergraduate level and OpenAI's graduate level on the 3rd: https://www.oneusefulthing.org/p/the-end-of-search-the-begin... And now this. I can't stop thinking about The O…

Mollik is the great sense-maker. Whatever he says is usually wrong

Re: Open Deep Research

#40

Post by swyx »

aubanel wrote: post #31

Hi all! Aymeric (m-ric) here, maintainer of smolagents and part of the team who built this. Happy to see this interesting people here! Few points: - open Deep Research is not a production app, but it could easily be productionized (would need to be faster + good UX). - As the GAIA score of 55% (not 54%, that would be lame) says, it's not far from the Deep Research score of 67%. It's also not there yet: I think the ma…

> for now we've used a text browser developed by the Microsofit autogen team, congrats to them

oh super cool! i've usually heard it the other way - people develop LLM-friendly web scrapers. i wrote one for myself, and for others there's firecrawl and expand.ai. a full "text browser" (i guess with rendering?) run locally seems like a better solution for local agents.

Post reply on HN