Updated
Updated
Items with an AI score under 20 are hidden. Show low-relevance items
Inferact@inferactOfficialAI score46AIInferact, with Red Hat AI, NVIDIA, and Huawei, co-led an optimization effort that makes Kimi K3 on vLLM 2.2–2.8× faster. The work spans scheduling, KDA state handling, and custom MoE kernels. The vLLM project's background post cites those throughput gains on a B300 benchmark against v0.27.1 and links a technical deep dive.
OpenBMB@OpenBMBOfficialAI score36AIOpenBMB's MiniCPM5-2B is a 2.5B-parameter model with native 128K context, offering hybrid Think and No-Think modes in one checkpoint. Users can download it from Hugging Face and run it fully offline on-device, as RunAnywhere demonstrated. In a demo, the model first called a puzzle impossible, then corrected itself and wrote a working verifier.
MiniMax (official)@MiniMax_AIOfficialAI score44AIVC-Attention, developed by Nunchux AI and collaborators, accelerates attention for MiniMax-H3 without retraining. In the B200 evaluation, it shows better fidelity than SageAttention2, with V-Smooth reducing value quantization error and ExpCast-FP8 speeding up softmax.
AIQianjue Technology founder Gao Haichuan argues that robotics will not see one breakthrough that suddenly lifts the whole industry, and he judges the company by deployment results rather than research papers. Qianjue, founded in 2023, has completed a Series A+ round worth a nine-figure RMB sum, with first orders coming from restaurant, cleaning, and hotel service robots. Gao says customers care about task completion, failure rates, and price rather than whether a predictive world model is used.
LM Studio@lmstudioOfficialAI score44AILM Studio's new Introspection feature lets its Bionic agent search its own session history, improving handling of long-term context across multiple compactions. Users can also reference other sessions directly in the composer with an @ mention.
AISkillsync, a Y Combinator W26 company, launched a tool that converts AI chat sessions between coding agents, including messages, reasoning and tool calls. The conversion engine txcript is open source, and the local-first app runs on Mac with a CLI and MCP support. Only sessions shared into team workspaces leave the user's machine.
Daniel Han@danielhanchenXAI score44AIUnsloth Desktop now supports multi-user accounts, alongside a revamped Docker image and custom Jupyter Notebook with custom themes, titles, and expandable cells. The update adds RDNA1+2 support, ARM64 Windows CUDA support, faster GRPO, and FP8/INT8 image diffusion support for 2x faster inference.
Yi Tay@YiTayMLXAI score38AIYi Tay congratulated Vinh Q. Tran (@vqctran) on his move from Google DeepMind and recalled years of co-inventing work together. The post praised Tran's pursuit of his biology passion, which the quoted post describes as building Polyphron to grow human tissue with frontier models and reinforcement learning.
Unsloth AI@UnslothAIOfficialPickAI score60AIUnsloth announced that its Docker image now lets users train and run more than 500 models locally with no setup required. The image works on NVIDIA and AMD hardware and supports a new GUI or notebook workflow. The post links to an installation guide and the GitHub repository.
Why it matters: The post explains how to run and train 500+ models locally with a no-setup Docker image, supported on NVIDIA and AMD, with a new GUI or notebook workflow.
Replicate@replicateOfficialAI score33AIPruna AI has introduced P-Video-2-Pro, built on MiniMax H3, which generates 5-second videos at 480p or 720p in seconds. The model is 50% off for one week, starting at $0.01 per second.
OpenBMB@OpenBMBOfficialAI score40AIOpenMed paired with MiniCPM5-2B to demonstrate a local clinical AI workflow combining privacy-preserving data processing with a compact model's tool use and long-context reasoning. OpenMed masks sensitive identifiers and extracts clinical context before MiniCPM5-2B calls tools, compares lab results, and generates clinical handoffs with source references. The post presents this as an example of keeping inference on local, resource-constrained hardware.
OpenBMB@OpenBMBOfficialAI score29AIDeveloper Alican Kiraz fine-tuned OpenBMB's open-source VoxCPM2 voice model on nearly 100 hours of natural Turkish speech, creating Kahya-TTS for Turkish text-to-speech. The project shows how open-source voice models can be adapted to new languages and specialized datasets. The model is available on Hugging Face.
OpenBMB@OpenBMBOfficialAI score20AIOpenBMB says Dubedo is the kind of product it hoped VoxCPM2 would enable, citing speaker-aware cloning, multilingual generation, and an editing studio. The post praises @dubedostudio's work, while the background post describes Dubedo as dubbing into 30 languages with per-speaker voice cloning and a beta open for trial.
Tinker@tinkerapiOfficialAI score32AISundial fine-tuned Thinking Machines' Inkling-Small with RLVR on 3,978 verified TeX.StackExchange fixes, using rewards for compilation and PDF match and penalties for removed content. The trained model fixes 83.7% of LaTeX errors in under one second at $0.0013 per fix, according to the post. Sundial says it is rolling out the model in its editor, applying fixes as suggestions and rebuilding the PDF.
Kilo (acq. by Anaconda)@kilocodeOfficialAI score40AIKilo Mobile now lets users spawn Cloud Agents, start sessions on remote machines, and dictate prompts by voice from a phone. Users can also review and comment on pull requests and approve Security Agent remediations without a laptop. On iPhone, Live Activities show session status on the Lock Screen when an agent needs input.
AIZed has launched the public beta of Delta, a multiplayer environment for coding with agents and reviewing their work, which replaces pull requests with shared threads. Delta is built on DeltaDB, which records edits and messages between Git commits, and it is free during the beta, with paid plans for individuals and teams to follow.
Why it matters: The post explains how Delta replaces pull requests with shared agent threads and DeltaDB, showing a concrete alternative to the GitHub review workflow.
Vercel@vercelOfficialAI score22AIDelphi, which turns experts' knowledge into digital minds, runs its Python backend on Vercel with a 10-person team and no dedicated infrastructure role. The stack uses Vercel Workflows for long-running agents and Vercel Queues for background jobs. The team reports more than 100 production deploys a day.
LM Studio@lmstudioOfficialAI score36AILM Studio has introduced local realtime voice transcription, letting users speak to their agents without sending voice data to the cloud. The feature is available on Mac and Windows, and now Linux as well.
AIAir Street Capital has led a $40 million Series A round in Jack & Jill, with Madrona joining and existing investors Creandum and Entrepreneurs First investing again. The company runs Jack, an AI agent that talks with job seekers about career moves and searches millions of job postings, and Jill, a recruiting agent that introduces candidates to hiring managers. Jack & Jill has arranged 25,000 interviews and plans 5,000 more each month, according to the article.
AIShanghai-based Twoo is building an AI that participates in conversations between two users, sharing context, remembering experiences, and helping them plan activities together rather than serving one person. Founder Cobe Chen says the product's monetization includes in-app "shells" that unlock outfits for its octopus AI character, plus premium memberships for professional uses such as collaborative writing and teaching. The company is preparing overseas expansion into North America, Japan, and South Korea.
Kilo (acq. by Anaconda)@kilocodeOfficialAI score22AIKilo announces that its Kilo App is live on Product Hunt, letting users start coding agents, check sessions, and review pull requests from iOS and Android. The company asks supporters to upvote or comment on its Product Hunt listing.
AINovita AI has open-sourced Chord, a W4A16 MoE CUDA operator with BF16 activations, INT4 weights and group-32 scales, built for Kimi K2.x serving shapes. Measured per layer against public Humming, it reports 1.11–1.20x on H200 EP8 prefill, 1.17–1.33x on H200 TP8 serving, and 1.81–2.15x on B300 EP8 decode against an untuned Humming default. Integration of the grouped operators with vLLM's Humming backend is still a work in progress.
Inferact@inferactOfficialAI score42AIInferact and Google Cloud announce a partnership to make Google TPUs a first-class platform in the vLLM open-source project. The collaboration targets production serving features, optimized kernels, a native PyTorch path via TorchTPU, and day-0 support for frontier model releases. A community program will offer shared TPU capacity and review and design help from vLLM core maintainers, with all outputs released as open source.
Intern Large Models@intern_lmOfficialAI score25AIIntern-S2-397B, a model built for long-horizon scientific research, now has Day-0 support in vLLM. The model brings multimodal, reasoning, coding, and scientific agent capabilities, and vLLM has published a run recipe for it.
Ian Johnson 🔬🤖@enjalotXAI score34AIIan Johnson visualizes 30 million jina-v5-nano embeddings from 12 billion tokens across multilingual FineWeb, StarCoder, The Pile, and RedPajama as a flyable video game. He frames the project as making data exploration engaging rather than a chore.
Inferact@inferactOfficialAI score34AIInferact says vLLM now supports DeepSeek v4.1 Flash on day zero across H100, H200, B200, B300, GB200, and GB300 GPUs. SemiAnalysis independently verified the NVIDIA support, while the post notes AMD vLLM still does not work with the model. Serving recipes are available at recipes.vllm.ai.
Vercel@vercelOfficialAI score26AITailscale offers instant access to hundreds of models for any user in a secure tailnet through its customer-facing model router, Aperture. Aperture is built on Vercel's AI Gateway and offers zero data retention, zero markup with free BYOK, and cost and usage data on every response.
LlamaIndex 🦙@llama_indexOfficialAI score22AILlamaParse's new high-effort mode provides granular page-level confidence scores with text explanations of parsing quality. The scoring also checks the original document as an extra verification step. High-effort mode costs 5 additional credits per page and is meant to be used only when needed.
ollama@ollamaOfficialAI score22AIOllama says users can enable ChatGPT inside its app, with downloads available at The post gives no further details on how the integration works or which models it covers.
ollama@ollamaOfficialAI score38AIOllama says users can customize which models appear in ChatGPT's model selector through settings. Both local and cloud models can be selected for display.
ollama@ollamaOfficialAI score22AIOllama states that plugins, web search, and computer use work out of the box. The post gives no further detail on which models, versions, or setup steps are involved.
ollama@ollamaOfficialAI score46AIThe ChatGPT Desktop app, referred to as the Codex app, can now be configured to run Ollama models. Users can download or update to Ollama 0.34 to get started.
Stephanie Palazzolo@steph_palazzoloXAI score36AIBill Peebles, OpenAI's former head of Sora, is teaming up with Hollywood mogul Jeffrey Katzenberg and investor Sujay Jaswa to create a new startup. The company will develop AI models for Hollywood, according to a report from The Information.
RadixArk@radixarkOfficialAI score22AIRadixArk has published a cookbook on its Miles documentation site covering how to run DeepSeek V4.1 Flash. The post itself contains only a link to the cookbook page, so no further details about features, figures, or setup steps are available.
AIBAAI released open-source resources for three AIDD projects on Hugging Face: EPT, an equivariant pretrained transformer for unified 3D molecular representation learning, and UniPath, a learnable-time flow matching method for crystal structure and energy prediction. The repository mirrors their GitHub source code and READMEs, with setup, preprocessing, training, and evaluation documentation. The MiSI benchmark is released separately on Hugging Face.
AIMireye, a Y Combinator S26 startup, launches an API and MCP server that supply AI agents with cited facts, property enrichment, tools, and change signals for any US location. The founder says a free tier of 5,000 credits is available with no card, and that his earlier site-screening app was dropped because customers wanted the underlying engine. Usage data cited in the post shows 311 of the 317 catalog fields are queried.
AINori Robotics, a Y Combinator S26 company, is selling the Nori A3, a humanoid robot priced at $1,688 with no deposit and shipping in fall 2026. The robot has 7+1 degree-of-freedom arms with 1.5 kg payload per arm, a 12 m lidar, four 720p RGB cameras, and 6-8 hours of battery life. Nori says it is assembled in San Francisco and offers a skills marketplace and a laptop app for training and operating the robot.
AIAlmanac, a Y Combinator S26 company, launches a personal AI workspace that brings together tasks, projects, conversations and a connected wiki. The workspace picks up follow-ups from connected email and calendar accounts, and the Mac desktop app is the starting point. A seven-day free trial requires a card, then the plan renews at $20 per month unless canceled, and model usage counts toward a connected ChatGPT subscription's Codex limits.