This product was not featured by Product Hunt yet. It will not be visible on their landing page and won't be ranked (cannot win product of the day regardless of upvotes).
Pinku
Private AI chat that runs offline on your iPhone and Mac
Pinku runs open models (Llama, Qwen, Gemma) on your iPhone, iPad and Mac with Apple's MLX. Download a model once and chat offline, even in airplane mode. No account, no server: your chats never leave your device. Free to start.
Hi Product Hunt! I built Pinku because I wanted to ask an AI things I wouldn't paste into a cloud chat (health questions, drafts, notes about people) without wondering where they end up.
Pinku runs the model on your own device with Apple's MLX framework. You pick from more than a dozen open models (from Qwen3 0.6B to Qwen3 8B), download one from Hugging Face, and from then on it works offline. There's no account and no server of ours that sees your chats. Pinku only goes online for model downloads, the list of available models, and Pro purchases. The privacy page lists every network call: https://aisagalab.com/pinku/priv...
What it's good at: quick drafts, rewriting, explaining, brainstorming, and offline use on flights. What it's not: a GPT-5 replacement. Small models are fast and private, but less knowledgeable.
It's free to start with Llama 3.2 1B. Pinku Pro (optional) unlocks the bigger models, custom instructions, on-device dictation and read-aloud.
I'd love to hear which models you want next and what you'd use a fully offline assistant for.
No comment highlights available yet. Please check back later!
About Pinku on Product Hunt
“Private AI chat that runs offline on your iPhone and Mac”
Pinku was submitted on Product Hunt and earned 0 upvotes and 1 comments, placing #57 on the daily leaderboard. Pinku runs open models (Llama, Qwen, Gemma) on your iPhone, iPad and Mac with Apple's MLX. Download a model once and chat offline, even in airplane mode. No account, no server: your chats never leave your device. Free to start.
Pinku was featured in Mac (103.8k followers), Privacy (11.5k followers) and Artificial Intelligence (482.8k followers) on Product Hunt. Together, these topics include over 157k products, making this a competitive space to launch in.
Who hunted Pinku?
Pinku was hunted by Marta. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how Pinku stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Marta
Hi Product Hunt! I built Pinku because I wanted to ask an AI things I wouldn't paste into a cloud chat (health questions, drafts, notes about people) without wondering where they end up.
Pinku runs the model on your own device with Apple's MLX framework. You pick from more than a dozen open models (from Qwen3 0.6B to Qwen3 8B), download one from Hugging Face, and from then on it works offline. There's no account and no server of ours that sees your chats. Pinku only goes online for model downloads, the list of available models, and Pro purchases. The privacy page lists every network call: https://aisagalab.com/pinku/priv...
What it's good at: quick drafts, rewriting, explaining, brainstorming, and offline use on flights. What it's not: a GPT-5 replacement. Small models are fast and private, but less knowledgeable.
It's free to start with Llama 3.2 1B. Pinku Pro (optional) unlocks the bigger models, custom instructions, on-device dictation and read-aloud.
I'd love to hear which models you want next and what you'd use a fully offline assistant for.