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

Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch

tongyi-agent.github.io

Post reply on HN
Print view View original

141–150 of 156 posts

Re: Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch

#141

Post by zo1 »

andai wrote: post #128
blaesus wrote: post #74

Earlier quoted context omitted.

"Summarization of what a search engine would return" is good enough for many of my purposes though. Good for breaking into new grounds, finding unknown unknowns, brainstorming etc.

I have a script that searches DDG (free), scrapes top 5 results, shoves them into an LLM, and answers your question. I wrote it back when AI web search was a paid feature and I wanted access to it. At the time Auto-GPT was popular and using the LLM itself to slowly and unreliably do the research. So I realized a Python program would be way faster and it would actually be deterministic in terms of doing what you expec…

If you think these things are just using a "dumb" search query, and using the top 5 hits, you're in for a lot of surprises very soon.

Re: Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch

#142

Post by whatpeoplewant »

Great to see an open 30B MoE aimed at “deep research.” These shine when used in a multi-agent setup: run parallel agentic AI workers (light models for browsing/extraction) and reserve the 30B agentic LLM for planning, tool routing, and verification—keeping latency/cost in check while boosting reliability. MoE specialization fits distributed agentic AI well, but you’ll want orchestration for retries/consensus and task-specific evals on multi-hop web research to guard against brittle routing and hallucinations.

Re: Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch

#143

Post by andai »

zo1 wrote: post #141
andai wrote: post #128

Earlier quoted context omitted.

I have a script that searches DDG (free), scrapes top 5 results, shoves them into an LLM, and answers your question. I wrote it back when AI web search was a paid feature and I wanted access to it. At the time Auto-GPT was popular and using the LLM itself to slowly and unreliably do the research. So I realized a Python program would be way faster and it would actually be deterministic in terms of doing what you expec…

If you think these things are just using a "dumb" search query, and using the top 5 hits, you're in for a lot of surprises very soon.

Well, considering TFA, it would be pretty strange if I did!

My point was it's silly to rely on a slow, expensive, unreliable system to do things you can do quickly and reliably with ten lines of Python.

I saw this in the Auto-GPT days. They tried to make GPT-4 (the non-agentic one with the 8k context window) use tool calls to do a bunch of tasks. And it kept getting confused and forgetting to do stuff.

Whereas if you just had

for page in pages: summarize(page)

it works 100% of the time, can be parallelized etc.

And of course the best part is that the LLM itself can write that code, i.e. it already has the power to make up for its own weaknesses, and make (parts of itself) run deterministically.

---

On that note, do you know more about the environment they ran this thing in? I got API access (it's free on OpenRouter), but I'm not sure what to plug this into. OpenRouter provides a search tool, but the paper mentions intelligent context compression and all sorts of things.

Re: Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch

#144

Post by threecheese »

tbruckner wrote: post #34

Has anyone found these deep research tools useful? In my experience, they generate really bland reports don't go much further than summarization of what a search engine would return.

Perplexity’s Research tool has basically replaced Google for me, for any search where I don’t already know the answer or know that it’s available somewhere (like documentation).

I use it dozens of times per day, and typically follow up or ask refining questions within the thread if it’s not giving me what I need.

It typically takes between 10sec and 5 minutes, and mostly replicates my manual process - search, review results, another 1..N search passes, review, etc. Initially it rephrases/refines my query, then builds a plan, and this looks a lot like what I might do manually.

Re: Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch

#146

Post by ainch »

pama wrote: post #93
ainch wrote: post #62

Earlier quoted context omitted.

Generally you train each expert simultaneously. The benefit of MoEs is that you get cheap inference because you only use the active expert parameters, which constitute a small fraction of the total parameter count. For example Deepseek R1 (which is especially sparse) only uses 1/18th of the total parameters per-query.

> only uses 1/18th of the total parameters per-query. only uses 1/18th of the total parameters per token. It may use the large fraction of them in a single query.

That's a good correction, thanks.

Re: Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch

#147

Post by Imustaskforhelp »

zurfer wrote: post #120
Imustaskforhelp wrote: post #99

Earlier quoted context omitted.

> But there is probably already some tradeoff, as GPT 3.5 was awesome at chess and current models don't seem trained extensively on chess anymore. Wow, I am so curious, can you provide me the source I am so interested in a chess LLM's benchmark as someone who occasionally plays chess. I have thought about creating things like these but it would be very interesting to find the best model at chess which isn't stockfish…

this was the article I had in mind, when writing this: https://dynomight.substack.com/p/chess

Ohhh I think this was the same article that I also had in mind

Key memory unlocked. I had an Aha moment with this article, thanks a lot for sharing it, appreciate it.

Re: Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch

#148

Post by mike_hearn »

MaxPock wrote: post #95
mike_hearn wrote: post #87

Earlier quoted context omitted.

It's certainly both a lot more than distillation and at least some Chinese labs have been cloning OpenAI via distillation. That's why they instituted much tighter ID verification requirements earlier this year. No, the reason you don't see many open source models coming from the rest-of-world (other than Mistral in France) is that you still need a ton of capital to do it. China can compete because the CCP used a comb…

There’s no Chinese lab that has been accused by OpenAI or anyone else of distillation. The accusations come from fringe right-wing media that are used to the “China only copies” trope. Training a model, by the way, is not about money, because many Western tech giants have more money than the CCP can allocate to Chinese labs. Apple, Meta, Amazon, SAP, IBM, and others have access to the same data as OpenAI and should t…

They didn't make a big fuss about it but OpenAI have explained that they instituted ID and country verification because there were competitors distilling their models. Of their competitors do you really think Anthropic, Google or Meta were doing that? It's pretty clear who they were talking about.

Chinese labs are mostly (all?) privately funded, as far as I know. Alibaba isn't a SOE. That's why I didn't mention state subsidies, although that might be happening (and certainly is happening w.r.t. access to electricity).

I didn't mention lax copyright/patent enforcement in the context of AI, but rather, the prior years in which China was able to build up local tech firms capable of taking on the US tech firms. It's mostly in the past now, they don't need to do that stuff anymore.

Re: Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch

#149

Post by threecheese »

ukuina wrote: post #126
oofbey wrote: post #104

Earlier quoted context omitted.

I’ve also had extremely poor luck getting any LLM agent to go through a long list of repetitive tasks. Don’t know why. I’d guess it’s because they’re trained for transactional responses, and thus are horrible at repute anything.

Very much this. You are better off asking it a write a script to invoke itself N times across the task list.

Same. I think there’s an untapped market (feature really) here, which if isn’t solved by GPT-next will start to reveal itself as a problem more and more.

LLMs are really bad at being comprehensive, in general, and from one inference to the next their comprehensive-ness varies wildly. Because LLMs are surprising the hell out of everyone with their abilities, less attention is paid to this; they can do a thing well, and for now that’s good enough. As we scale usage, I expect this gap will become more obvious and problematic (unless solved in the model, like everything else).

A solution I’ve been toying with is something like a reasoning step, which could probably be done with mostly classical NLP, that identifies constraints up front and guides the inference to meet them. Like a structured output but at a session level.

I am currently doing what you suggest though, I have the agent create a script which invokes … itself … until the constraints are met, but that obviously requires that I am engaged there; I think it could be done autonomously, with at least much better consistency (at the end of the day even that guiding hand is inference based and therefore subject to the same challenges).

Post reply on HN