🔥 sponsors/dmtrKovalenko
Hi there, little explorer! 👋
Imagine you have a super messy toy box, and you're looking for your favorite red car. 🚗💨
This news is about a special helper, like a super-fast robot friend, that can find things in a giant toy box (which is like a computer!) super-duper quickly! It's like magic! ✨
This robot friend helps other smart robots (we call them AI agents) find their toys (which are like important computer pieces) super fast. So they can play and learn even quicker!
It's a cool new tool that helps computers be even smarter and faster. Yay! 🎉
The fastest and the most accurate file search toolkit for AI agents, Neovim, Rust, C, and NodeJS — Trending on GitHub today with 767 new stars.
So you are here because one of my projects was helpful to you. I am very happy if this is true. You can support my open-source work here and I will appreciate any donation!
For sure, if you are sponsoring me your issues are prioritized automatically and I will try to implement your feature requests much faster.
If you are stuck with some issue and need immediate help — you can donate and contact me at [email protected] and I will help you to resolve your issue ASAP.
Here are the projects I am currently maintaining:
odiff
Link: https://github.com/dmtrKovalenko/odiff What it is: A pretty fast image comparison tool
It already powers several visual regression services and different tools that increase the speed of visual regression testing. And it can be even faster and even more helpful with your help.
cypress-real-events
Link: https://github.com/dmtrKovalenko/cypress-real-events What it is: Native system events plugin for Cypress
As far as I know, this plugin literally saves people's test suites! Super proud of this project because it unlocks features that have never been possible in Cypress before, like hovering, native clicks, keypress, drag and drops, and many more.
date-io
Link: https://github.com/dmtrKovalenko/date-io What it is: Adapter across different javascript date libraries
This project is huge! It powers a lot of giant projects like https://github.com/dmtrKovalenko/odiff Material-UI date pickers (which to be honest were also written by me but a very long time ago). It has altogether 15 million downloads per month and helps maintain to build useful date & time controls. And you can help make all of them even better!
blendr
Link: https://github.com/dmtrKovalenko/blendr What it is: Bluetooth low energy browser terminal app
This project is relatively new but already used by several full-time bluetooth engineering teams because it saves huge amount of time by automating debug process of BLE devices. It is a vim-style terminal app to connect, explore, and debug BLE devices around with 2 killer features.
fframes
Link: http://fframes.studio What it is: Video rendering framework in Rust
This project is still in beta. But it is enormous! End-to-end rust video rendering engine that is trying to be as fast as possible by optimizing everything from markup parsing to encoding the rendered frames. By doing so it will be possible to render beautiful motion graphics videos on a CPU in a reasonable amount of time. Well, if I will finish it. And you know, heh, you can help me to make this happen!
Thanks for reading and for being willing to support my work. I will appreciate any donation cause it will help me to spend more time doing my favorite job.
Have a nice day!
Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
githubtrendingopen-source
LLM-Agent-based Social Simulation for Attitude Diffusion
arXiv:2604.03898v1 Announce Type: new Abstract: This paper introduces discourse_simulator, an open-source framework that combines LLMs with agent-based modelling. It offers a new way to simulate how public attitudes toward immigration change over time in response to salient events like protests, controversies, or policy debates. Large language models (LLMs) are used to generate social media posts, interpret opinions, and model how ideas spread through social networks. Unlike traditional agent-based models that rely on fixed, rule-based opinion updates and cannot generate natural language or consider current events, this approach integrates multidimensional sociological belief structures and real-world event timelines. This framework is wrapped into an open-source Python package that integr
Affording Process Auditability with QualAnalyzer: An Atomistic LLM Analysis Tool for Qualitative Research
arXiv:2604.03820v1 Announce Type: new Abstract: Large language models are increasingly used for qualitative data analysis, but many workflows obscure how analytic conclusions are produced. We present QualAnalyzer, an open-source Chrome extension for Google Workspace that supports atomistic LLM analysis by processing each data segment independently and preserving the prompt, input, and output for every unit. Through two case studies -- holistic essay scoring and deductive thematic coding of interview transcripts -- we show that this approach creates a legible audit trail and helps researchers investigate systematic differences between LLM and human judgments. We argue that process auditability is essential for making LLM-assisted qualitative research more transparent and methodologically ro
Three Phases of Expert Routing: How Load Balance Evolves During Mixture-of-Experts Training
arXiv:2604.04230v1 Announce Type: cross Abstract: We model Mixture-of-Experts (MoE) token routing as a congestion game with a single effective parameter, the congestion coefficient gamma_eff, that quantifies the balance-quality tradeoff. Tracking gamma_eff across training checkpoints of two open-source MoE models, OLMoE-1B-7B (20 checkpoints, with dense sampling in the surge region) and OpenMoE-8B (6 checkpoints), reveals a three-phase trajectory: a surge phase where the router learns to balance load (gamma_eff: 14 to 36-39, peaking in the step 30K-40K region), a stabilization phase where experts specialize under steady balance (B_0: 2.4 to 2.3, steps 100K-400K), and a relaxation phase where the router trades balance for quality as experts differentiate (gamma_eff: 27 to 9, steps 400K-1.2M
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Open Source AI
Optimizing Neurorobot Policy under Limited Demonstration Data through Preference Regret
arXiv:2604.03523v1 Announce Type: new Abstract: Robot reinforcement learning from demonstrations (RLfD) assumes that expert data is abundant; this is usually unrealistic in the real world given data scarcity as well as high collection cost. Furthermore, imitation learning algorithms assume that the data is independently and identically distributed, which ultimately results in poorer performance as gradual errors emerge and compound within test-time trajectories. We address these issues by introducing the "master your own expertise" (MYOE) framework, a self-imitation framework that enables robotic agents to learn complex behaviors from limited demonstration data samples. Inspired by human perception and action, we propose and design what we call the queryable mixture-of-preferences state sp
langchain-ollama==1.1.0
Changes since langchain-ollama==1.0.1 release(ollama): 1.1.0 ( #36574 ) feat(ollama): support response_format ( #34612 ) fix(ollama): serialize reasoning_content back to ollama thinking ( #36573 ) fix(ollama): prevent _convert_messages_to_ollama_messages from mutating caller list ( #36567 ) feat(ollama): add dimensions to OllamaEmbeddings ( #36543 ) fix(ollama): respect scheme-less base_url ( #34042 ) feat(ollama): logprobs support in Ollama ( #34218 ) chore(ollama): switch to ty ( #36571 ) chore: add comment explaining pygments>=2.20.0 ( #36570 ) chore: pygments>=2.20.0 across all packages ( CVE-2026-4539 ) ( #36385 ) chore: bump requests from 2.32.5 to 2.33.0 in /libs/partners/ollama ( #36249 ) chore(partners): bump langchain-core min to 1.2.21 ( #36183 ) ci: suppress pytest streaming ou
[llama.cpp] 3.1x Q8_0 speedup on Intel Arc GPUs - reorder optimization fix (PR submitted)
TL;DR : Q8_0 quantization on Intel Xe2 (Battlemage/Arc B-series) GPUs was achieving only 21% of theoretical memory bandwidth. My AI Agent and I found the root cause and submitted a fix that brings it to 66% - a 3.1x speedup in token generation. The problem : On Intel Arc Pro B70, Q8_0 models ran at 4.88 t/s while Q4_K_M ran at 20.56 t/s; a 4x gap that shouldn't exist since Q8_0 only has 1.7x more data. After ruling out VRAM pressure, drivers, and backend issues, we traced it to the SYCL kernel dispatch path. Root cause : llama.cpp's SYCL backend has a "reorder" optimization that separates quantization scale factors from weight data for coalesced GPU memory access. This was implemented for Q4_0, Q4_K, and Q6_K - but Q8_0 was never added. Q8_0's 34-byte blocks (not power-of-2) make the non-r
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!