All activity
Jim Engineleft a comment
Okay but which languages? Why not show the 10 languages more obvious
Qwen3-TTSVoice design, cloning & 97ms streaming
Jim Engineleft a comment
It's good OCR model but they have increased prices by 100% for this. If they keep doing that, this will not be really positive for their business and they will lose a lot of customers even though their model is good
Mistral OCR 3Accurate OCR for notes, forms, tables and handwriting
xPrivo is a free, open-source, private AI chat assistant that focuses entirely on keeping you anonymous. You never need to create an account, even with the PRO membership. You can either self-host it or use it directly via the website. xPrivo uses modern, powerful open-source models such as Mistral 3 and DeepSeek V3.2, which are self-hosted in the EU. Chats are never logged or used for training, and they stay completely local to your device. Your chat history is never stored elsewhere.
xPrivoOpen Source, Free Anonymous AI Chat - Ready to Run Locally
Jim Engineleft a comment
By default, the most powerful AI assistants train their new models using your conversations. They can also read your personal chats with the AI assistants. With xPrivo, however, you have the power to keep your chats completely private and ensure they remain on your device. As it is open source, you can even run it on your own device and have an AI assistant that runs entirely locally, provided...
xPrivoOpen Source, Free Anonymous AI Chat - Ready to Run Locally
Jim Engineleft a comment
I think the core is a good idea but why SMS, it could also be just via SMTP mail and litereally cost nothing? SMS can become quite expensive if you have a bigger side project where a lot of messages are sent then. I would stick to email and send it to an email address that has priority on your phone so it notifies you like a SMS would do. Maybe I'm wrong but that are my 2 cents
Console.text()SMS alerts for your code just like console.log()
Jim Engineleft a comment
Yeah what should I say. It's another Mistral release within 1 week and I've tested it yesterday. Must admit that it's really good and the Devstrall Small is punching above it's weight. Works smooth on my M4 chip and can't complain yet 👌🏻
Devstral 2SOTA open-source agentic coding models and CLI agent
Jim Engineleft a comment
Alright so the dashboard looks good with the routine creation I must admit. Also was in the same situation as you with existing browser agents missing things or being not capable of what I wanted so your solution might solve this now.
VectorlyGet web data behind clicks, searches, user interactions
Jim Engineleft a comment
Are there any known names in the files or are they all redacted which would then make your AI say that there are no known name in it? How do you handle this?
EpsteinGPTChatGPT connected to the Epstein Files
Jim Engineleft a comment
Their progress is amazing and I actually love them now more than I loved ChatGPT when it came out. Somehow ChatGPT, even hough they improved GPT models seem to stagnate whereas Z.ai really shines with their Open Source releases. (Which also shows that OpenAI should actually have stayed open, I am pretty sure they would rock the generative AI world by now)
GLM-4.6VOpen-source multimodal model with native tool use
Jim Engineleft a comment
Nice that you've listened and improved the landing page a bit. Shows that you are really working hard on this one. I also like the idea of the visual first AI prototyping. @nathanngai
ArkhetAI prototyping with wireframes, not prompts
Jim Enginestarted a discussion
Are product launches and hunts increasing on PH since we have generative AI and vibe coding?
I am wondering if it's just me or if there are many more products to browse here on Producthunt every day since Vibe Coding joined the chat. I have been here for quite a few years and have seen lots of launches and hunts. But over the last one to two years, I have seen many more products being launched per day. So also more time needed for me to find good products. Maybe the Producthunt staff...
Jim Engineleft a comment
Love it, as from time to time one has to convert images and when googling you will always land on shady websites where you don't want to upload your personal images. I will use your image converter in the future for that reason. Any plans to open source it? Would be nice too
TransfiguratorFast, private image converter - 100% local processing on Mac
Jim Engineleft a comment
Okay a lot of informations on the website. But where is it hosted? Is it wrapped around Goolge Cloud, AWS or do your have your own datacenters with the GPUs. I don't really get it.
Inference Engine by GMI CloudFast multimodal-native inference at scale
Jim Engineleft a comment
Love the retro design, also that it's Open Source. Knew immediately it has to do something with shadcn 🫡
8bitcn/ui8‑bit UI components that work in any framework
Jim Engineleft a comment
So not yet for Pro users? I would honestly feel betrayed if I were paying for Pro and would like to test this new feature
Gemini 3 Deep Think by GoogleGoogle’s best model for logical thinking and understanding
Jim Engineleft a comment
Almost thought this would be a launch of Google's coding agent Antigravity 😅 But damn I realy want an FPV drone and feel like I'm flying myself through my neighbourhood but pricing is unfortunately too big for my budget. I hope that these type of drones could go down to like $300 and I would be so hyped. But still, this drone looks sick, also with the 8k 360 video ability
AntigravityThe world's first 8K 360° drone
Jim Engineleft a comment
Your landing page is partly broken
UnosendSend transactional + marketing email w/ 99.9% deliverability
Jim Engineleft a comment
Just tested it and it's realy good. I still think google Nano Banana Pro is better with more realistic images and multiple people in an image. But price-wise Seedream 4.5 beats Banana Pro and they are trying to really close the gap to google's SOTA image model.
Seedream 4.5High-fidelity multi-image editing & dense text rendering
Jim Engineleft a comment
Very nice, with Openrouter & co you'd not be able to bring your own key
TrueFoundry AI GatewayConnect, observe & control LLMs, MCPs, Guardrails & Prompts
Jim Engineleft a comment
I personally tried Mistral 3 Large when it came out yesterday and think it's a good model in terms of multilanguage and multimodality (as even the smaller models with less params can run purely in browser and confidently detect objects correctly - so kudos for that). But the Mistral 3 Large itself still seems to need a bit more power as in my own evaluations it still lags a bit behind the new...
Mistral 3A family of frontier open-source multimodal models