Making AI chat sessions durable to network failures
The experience with AI chat apps can be a bit fragile — there's no resilience to network failures, and responses aren't deterministic like traditional search. As Garry Tan put it (https://x.com/garrytan/status/1927038513108701662), "It feels like catastrophic data loss when a given response comes back and fails halfway, and then the next retry is not as good." Vercel recently released a "resumable-stream" package (https://github.com/vercel/resumable-stream) to solve this problem using Redis as a backing store. I felt that we could simplify the…
What it does
In the maker’s words, at launch
The experience with AI chat apps can be a bit fragile — there's no resilience to network failures, and responses aren't deterministic like traditional search. As Garry Tan put it (https://x.com/garrytan/status/1927038513108701662), "It feels like catastrophic data loss when a given response comes back and fails halfway, and then the next retry is not as good." Vercel recently released a "resumable-stream" package (https://github.com/vercel/resumable-stream) to solve this problem using Redis as a backing store. I felt that we could simplify the approach using s2.dev (I'm an engineer on the team), and created a fork which ended up borrowing the API but otherwise completely rewritten: https://github.com/s2-streamstore/resumable-stream - S2 streams are cheap, so instead of optimizing to flush only when there is another reader, the stream is always created and written to. - Unlike the typically ephemeral nature of Redis, tokens are always durable, and there are no memory constraints around number of streams or how much data is on them. - Relies on S2’s built-in concurrency control. A unique fencing token ensures a single writer, and a sequence number is optimistically supplied to match the one expected to be assigned to the next record, to deduplicate in case of retries. It integrates nicely with Chat SDK, and you can try this out here: https://ai-chatbot-s2.vercel.app by reloading the page during a larger response! There are some interesting future possibilities, like building a multiplayer chat experience around shared streams.
Does the same job
all alternatives →
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, August 2025
the whole month →
- IS
I built the world's most impractical 1000-pixel display and anyone in the world can draw on it. It draws a single pixel at a time and takes 30-60 minutes to complete a single image. Anyone can participate in the project by voting for the next image to be drawn, and submitting images. https://kilopx.com/
Work · 2025 · benholmen.com
- KT
Kitten TTS is an open-source series of tiny and expressive text-to-speech models for on-device applications. We are excited to launch a preview of our smallest model, which is less than 25 MB. This model has 15M parameters. This release supports English text-to-speech applications in eight voices: four male and four female. The model is quantized to int8 + fp16, and it uses onnx for runtime. The model is designed to run literally anywhere eg. raspberry pi, low-end smartphones, wearables, browsers etc. No GPU required! We're releasing this to give early users a sense of the latency and voices…
Dev tools · 2025 · github.com
- IW
I was wondering how I can arrange objects along a spherical helix path, and read some articles on it. I ended up learning about parametric equations again, and make this visualization to document what I learned: https://visualrambling.space/moving-objects-in-3d/ feel free to visit and let me know what you think!
Life & fun · 2025 · visualrambling.space
- TC
For HTML Day 2025 [1], I made a web service that displays the current sky at your approximate location as a CSS gradient. Colours are simulated on-demand using atmospheric absorption and scattering coefficients. Updates every minute, without the use of client-side JavaScript. Source code and additional information is available on GitHub: https://github.com/dnlzro/horizon [1] https://html.energy/html-day/2025/index.html
Dev tools · 2025 · sky.dlazaro.ca