Isn't OpenIA "Deep research" (not "DeepResearch") a methodology/tooling thing, and you'll get different responses depending on what specific model you use with it? As far as the UI allows you to, you could use Deep research with GPT-5, GPT-4o, o3 and so on, and that'll have an impact on the responses. Skimming the paper and searching for some simple terms makes it seem like they never expand on what exact models they've used, just that they've used a specific feature from ChatGPT?
At this point "deep research" is more of a pattern - OpenAI and Perplexity and Google Gemini all offer products with that name which work essentially the same way, and Anthropic and Grok have similar products with a slightly different name attached.
The pattern is effectively long-running research tasks that drive a search tool. You give them a prompt, they churn away for 5-10 minutes running searches and they output a report (with "citations") at the end.
This Tongyi model has been fine-tuned to be really good at using its search tool in a loop to produce a report.
Sunday morning, and I find myself wondering how the engineering tinkerer is supposed to best self-host these models? I'd love to load this up on the old 2080ti with 128gb of vram and play, even slowly. I'm curious what the current recommendation on that path looks like.
Constraints are the fun part here. I know this isn't the 8x Blackwell Lamborghini, that's the point. :)
It makes me wonder if we'll see an explosion of purpose trained LLMs because we hit diminishing returns on invest with pre training or if it takes a couple of months to fold these advantages back into the frontier models.
Given the size of frontier models I would assume that they can incorporate many specializations and the most lasting thing here is the training environment.
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.
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.
I made a 4B Qwen3 distill of this model (and a synthetic dataset created with it) a while back. Both can be found here: https://huggingface.co/flashresearch
I recently got a 5090 with 64 GB of RAM (intel cpu). Was just looking for a strong model I can host locally. If I had performance of GPT4-o, I'd be content. Are there any suggestions or cases where people got disappointed?
In my experience using these supposed expert models, they are all more or less the same given they all are trained on the same internet data. The differentiation and value is in the context window management and how relevant info from your session is pulled in. So it’s the interface to the model that makes all the difference. Even there the differences are quite minimal. That is because all these companies want to toe the line between providing functionality to keep the users engaged and pushing them to sign up for the subscription.
All this to ask the question, if I host these open source models locally, how is the user interface layer that remembers and picks the right data from my previous session and the agentic automation and others implemented? Do I have to do it myself or are the free options for that?
Is China dominating the US in terms of AI? Given that they currently have a model that beats the best models at all formal quantitative benchmarks?
What is the state of AI in China? My personal feeling is that it doesn't dominate the zeitgeist in China as it does in the US and despite this because of the massive amount of intellectual capital they have just a small portion of their software engineering talent working on this is enough to go head to head with us even though it only takes a fraction of their attention.
Recently I gave a list of 300 links to deep research and asked it to go through each one to analyze a certain question about them. Repeatedly it would take shortcuts and not actually do the full list. Is this caused by a context window limits? Or maybe Open AI limits request size? Is it possible to not run into these types of limits with locally hosted models?
For most plans, Deep Research is capped at around 20 sources, making it for many cases the least useful research agent, in particular worse than a thinking mode Gpt5 query
I tied it together with qwen3 30b thinking. Very easy to get it up and running, but lots of the numbers are shockingly low. You need to boost iterations and context. Especially easy if you already run searxng locally.
I havent finished tuning the actual settings, but for the detailed report it'll take ~20 minutes and so far has given pretty good results. Similar to openai's deep research. Mine often has ~100 sources.
But something I have noticed. It didnt seem to me the model was important. The magic was moreso in the project. Getting deep with higher iterations and more results.
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.
30 comments
[ 2.6 ms ] story [ 68.7 ms ] threadThe pattern is effectively long-running research tasks that drive a search tool. You give them a prompt, they churn away for 5-10 minutes running searches and they output a report (with "citations") at the end.
This Tongyi model has been fine-tuned to be really good at using its search tool in a loop to produce a report.
Constraints are the fun part here. I know this isn't the 8x Blackwell Lamborghini, that's the point. :)
Given the size of frontier models I would assume that they can incorporate many specializations and the most lasting thing here is the training environment.
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.
I switch between gemini and ChatGpt whenever I feel one fails to fully grasp what I want, I do coding in claude.
How are they supposed to become the 1 trillion dollar company they want to be, with strong competition and open source disruptions every few months?
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.
All this to ask the question, if I host these open source models locally, how is the user interface layer that remembers and picks the right data from my previous session and the agentic automation and others implemented? Do I have to do it myself or are the free options for that?
What is the state of AI in China? My personal feeling is that it doesn't dominate the zeitgeist in China as it does in the US and despite this because of the massive amount of intellectual capital they have just a small portion of their software engineering talent working on this is enough to go head to head with us even though it only takes a fraction of their attention.
For most plans, Deep Research is capped at around 20 sources, making it for many cases the least useful research agent, in particular worse than a thinking mode Gpt5 query
https://openrouter.ai/alibaba/tongyi-deepresearch-30b-a3b
https://openrouter.ai/alibaba/tongyi-deepresearch-30b-a3b:fr...
I tied it together with qwen3 30b thinking. Very easy to get it up and running, but lots of the numbers are shockingly low. You need to boost iterations and context. Especially easy if you already run searxng locally.
I havent finished tuning the actual settings, but for the detailed report it'll take ~20 minutes and so far has given pretty good results. Similar to openai's deep research. Mine often has ~100 sources.
But something I have noticed. It didnt seem to me the model was important. The magic was moreso in the project. Getting deep with higher iterations and more results.