Faster FastAPI with simdjson and io_uring on Linux 5.19
A few months ago, I benchmarked FastAPI on an i9 MacBook Pro. I couldn't believe my eyes. A primary REST endpoint to `sum` two integers took 6 milliseconds to evaluate. It is okay if you are targeting a server in another city, but it should be less when your client and server apps are running on the same machine. FastAPI would have bottleneck-ed the inference of our lightweight UForm neural networks recently trending on HN under the title "Beating OpenAI CLIP with 100x less data and compute". (Thank you all for the kind words!) So I wrote another library. It has been a while since I have…
In plain words
This library accelerates FastAPI applications on Linux 5.19 by leveraging simdjson for faster JSON parsing and io_uring for improved network I/O performance. It targets developers building machine learning inference servers or other latency-sensitive REST APIs where local request-response times matter. The library uses pre-registered buffers and file descriptor pooling to reduce overhead compared to standard FastAPI implementations.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
A few months ago, I benchmarked FastAPI on an i9 MacBook Pro. I couldn't believe my eyes. A primary REST endpoint to `sum` two integers took 6 milliseconds to evaluate. It is okay if you are targeting a server in another city, but it should be less when your client and server apps are running on the same machine. FastAPI would have bottleneck-ed the inference of our lightweight UForm neural networks recently trending on HN under the title "Beating OpenAI CLIP with 100x less data and compute". (Thank you all for the kind words!) So I wrote another library. It has been a while since I have written networking libraries, so I was eager to try the newer io_uring networking functionality added by Jens Axboe in kernel 5.19. TLDR: It's excellent! We used pre-registered buffers and re-allocated file descriptors from a managed pool. Some other parts, like multi-shot requests, also look intriguing, but we couldn't see a flawless way to integrate them into UJRPC. Maybe next time. Like a parent with two kids, we tell everyone we love Kernel Bypass and SIMD equally. So I decided to combine the two, potentially implementing one of the fastest implementations of the most straightforward RPC protocol - JSON-RPC. ~~Healthy and Fun~~ Efficient and Simple, what can be better? By now, you may already guess at least one of the dependencies - `simdjson` by Daniel Lemiere, that has become the industry standard. io_uring is generally very fast, even with a single core. Adding more polling threads may only increase congestion. We needed to continue using no more than one thread, but parsing messages may involve more work than just invoking a JSON parser. JSON-RPC is transport agnostic. The incoming requests can be sent over HTTP, pre-pended by rows of headers. Those would have to be POSTs and generally contain Content-Length and Content-Type. There is a SIMD-accelerated library for that as well. It is called `picohttpparser`, uses SSE, and is maintained by H2O. The story doesn't end there. JSON is limited. Passing binary strings is a nightmare. The most common approach is to encode them with base-64. So we took the Turbo-Base64 from the PowTurbo project to decode those binary strings. The core implementation of UJRPC is under 2000 lines of C++. Knowing that those lines connect 3 great libraries with the newest and coolest parts of Linux is enough to put a smile on my face. Most people are more rational, so here is another reason to be cheerful. - FastAPI throughput: 3'184 rps. - Python gRPC throughput: 9'849 rps. - UJRPC throughput: -- Python server with io_uring: 43'000 rps. -- C server with POSIX: 79'000 rps. -- C server with io_uring: 231'000 rps. Granted, this is yet to be your batteries-included server. It can't balance the load, manage threads, spell S in HTTPS, or call parents when you misbehave in school. But at least part of it you shouldn't expect from a web server. After following the standardization process of executors in C++ for the last N+1 years, we adapted the "bring your runtime" and "bring your thread-pool" policies. HTTPS support, however, is our next primary objective. --- Of course, it is a pre-production project and must have a lot of bugs. Don't hesitate to report them. We have huge plans for this tiny package and will potentially make it the default transport of UKV: https://github.com/unum-cloud/ukv
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, March 2023
the whole month →
- BI
I'm a big fan of the BBC podcast In Our Time -- and (like most people) I've been playing with the OpenAI APIs. In Our Time has almost 1,000 episodes on everything from Cleopatra to the evolution of teeth to plasma physics, all still available, so it's my starting point to learn about most topics. But it's not well organised. So here are the episodes sorted by library code. It's fun to explore. Web scraping is usually pretty tedious, but I found that I could send the minimised HTML to GPT-3 and get (almost) perfect JSON back: the prompt includes the Typescript definition. At the same time I…
AI · 2023 · genmon.github.io