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

Moondream 3 Preview: Frontier-level reasoning at a blazing speed

moondream.ai

Post reply on HN
Print view View original

11–20 of 46 posts

Re: Moondream 3 Preview: Frontier-level reasoning at a blazing speed

#11

Post by radq »

scoots_k wrote: post #7

Moondream 2 has been very useful for me: I've been using it to automatically label object detection datasets for novel classes and distill an orders of magnitude smaller but similarly accurate CNN. One oddity is that I haven't seen the claimed improvements beyond the 2025-01-09 tag - subsequent releases improve recall but degrade precision pretty significantly. It'd be amazing if object detection VLMs like this repor…

Thanks! If you could shoot me a note at vik@m87.ai with any examples of the precision/recall issues you saw I'd appreciate it a ton.

Re: Moondream 3 Preview: Frontier-level reasoning at a blazing speed

#12

Post by radq »

sheepscreek wrote: post #10

Impressive stuff! Has anyone tried it for computer/browser control? How does it fare with graphs and charts?

The 'point' skill is trained on a ton of UI data; we've heard of a lot of people using it in combination with a bigger driver model for UI automation. We are also planning on post-training it to work end-to-end for this in an agentic setting before the final release -- this was one of the main reasons we increased the model's context length.

Re: chart understanding, there are a lot of different types of charts out there but it does fairly well! We posted benchmarks for ChartQA in the blog but it's on par with GPT5* and slightly better than Gemini 2.5 Flash.

* To be fair to GPT5, it's going to work well on many more types of charts/graphs than Moondream. To be fair to Moondream, GPT5 isn't really well suited to deploy in a lot of vision AI applications due to cost/latency.

Re: Moondream 3 Preview: Frontier-level reasoning at a blazing speed

#15

Post by robertdaniels »

It's ability to process large volumes of images with low active parameters makes it a significant advancement for edge devices. However, scaling these models to production environments often introduces security challenges, including bot floods targeting inference APIs and adversarial inputs that mimic legitimate queries to disrupt detections.

Re: Moondream 3 Preview: Frontier-level reasoning at a blazing speed

#16

Post by scoots_k »

radq wrote: post #11
scoots_k wrote: post #7

Moondream 2 has been very useful for me: I've been using it to automatically label object detection datasets for novel classes and distill an orders of magnitude smaller but similarly accurate CNN. One oddity is that I haven't seen the claimed improvements beyond the 2025-01-09 tag - subsequent releases improve recall but degrade precision pretty significantly. It'd be amazing if object detection VLMs like this repor…

Thanks! If you could shoot me a note at vik@m87.ai with any examples of the precision/recall issues you saw I'd appreciate it a ton.

Will do!
Post reply on HN