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

VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO

arxiv.org

Post reply on HN
Print view View original

151–160 of 226 posts

Re: VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO

#151

Post by SubiculumCode »

brainless wrote: post #67

I recently came across this model and I would love to try it with my coding agent soon. I really like the idea of small models that can reason but do not have too much knowledge. Also, no emphasis on tool calls. I think the agent should do the heavy lifting and reach half way. I use really small models, like Qwen 3.5 0.8B to 9B - no tool calling, no MCP, no skills, nothing. No multi-turn chat even. Models are given v…

Maybe no tool calling, but seems it could be really good at deciding which tool to use and when?

Re: VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO

#152

Post by cogman10 »

hypendev wrote: post #144
secretslol wrote: post #23

Am I right in thinking this is a tiny model which has been trained well to reason, and that's it? Makes me think of a smart person who doesn't know anything about a given topic, but with the right tools will go and research the heck out of it. I really like the sound of this... why have models train on learning anything when you can just train them how to learn and let them get on with it from something as small as a…

Because reasoning is an emergent byproduct of training it on all knowledge. It still doesn't "know" things in this form and just generates tokens, no matter how weird we spin it. So if you don't train it on a large dataset of a lot of words with a lot of sensible connections, it won't be able to reason, as it won't be able to make proper connections between words and sentences. You can try training a really small mod…

I think small expert models could be pretty powerful from open weight providers.

Imagine, for example, a model that's primarily train on typescript and general programming. It would be faster to train and it could be a lot smaller than a generalist model. It might be the best model to pick when you are doing typescript programming. And if you could squeeze that into 3B parameters a lot of consumer hardware could run it locally.

You could even expand it to just "webdev tech" or the like.

Re: VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO

#153

Post by yousif_123123 »

I really hope that in a couple of years I can have a laptop that runs a reasonably good coding agent locally, that I can run fast and do most of my programming with, without running my laptop hot. I could keep open code and use other models when needed, but really for most of my work, I'm already breaking it down so that I can review code changes eventually, and I just need something reasonably decent and fast and unlimited. I think its coming.

Re: VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO

#154

Post by YetAnotherNick »

deftio wrote: post #16

There is some base level of intelligence any model needs to be useful, even in narrow tasks. Could you teach a 5 year old to drive a car? A 10 year old? A 12 year old? To drive a car requires being able to read, to have judgement about ice or rainy conditions, to anticipate a child running after a ball. By the time a human in in their mid teens they have acquired the base knowledge... Small models need to have enough…

> A 10 year old

https://www.youtube.com/watch?v=sLIAoW4QxIs

Re: VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO

#156

Post by alkonaut »

yousif_123123 wrote: post #153

I really hope that in a couple of years I can have a laptop that runs a reasonably good coding agent locally, that I can run fast and do most of my programming with, without running my laptop hot. I could keep open code and use other models when needed, but really for most of my work, I'm already breaking it down so that I can review code changes eventually, and I just need something reasonably decent and fast and un…

I hope so too. But I fear that it will feel inadequate if we know there is always a $20 online model that is an order of magnitude better. I don't think there will be a "good enough" local model so long as frontier models look so much better.

Re: VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO

#158

Post by yogthos »

yousif_123123 wrote: post #153

I really hope that in a couple of years I can have a laptop that runs a reasonably good coding agent locally, that I can run fast and do most of my programming with, without running my laptop hot. I could keep open code and use other models when needed, but really for most of my work, I'm already breaking it down so that I can review code changes eventually, and I just need something reasonably decent and fast and un…

I'm very optimistic here as well. And it's also worth noting that tooling is improving along with the models. I really think we have to treat models and tools as a package. The models is your engine, but you need a chassis to run it.

I find what makes frontier models actually work well isn't just the capability of the model, but how well the harness is tuned to its expectations. I wrote a about this in a bit more detail here. https://yogthos.net/posts/2026-06-08-dirge-code.html

Re: VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO

#160

Post by bmacho »

dandaka wrote: post #133
XCSme wrote: post #102

Earlier quoted context omitted.

Yup, you still need knowledge. Even if you have access to all the data and tools, you still need to know what to search for, what tools to use and to understand what the user is asking. Our computers can already do everything, have access to all the tools and information, yet they still need a human/intelligence to use it and apply to specific problems. Even defining the problem requires knowledge. As for the tools,…

Model can use tools to get that knowledge. In your example, read Wikipedia page about table tennis. Imagine a reasoning engine with a big enough context, that knows nothing. A path built from first principles to understand "table tennis spin" — does not look very long for me.

It probably can't use all the truth in its content window, not yet anyway.

E.g. you put a graph in its content window, and you ask it to find a Hamiltonian cycle, can it do it?

Probably this could be a next step in the future for more powerful AIs, a layer that abstracts the facts in its content window away, and a layer that solves this types of abstractions.

Post reply on HN