Updated
Updated
Items with an AI score under 20 are hidden. Show low-relevance items
Jerry Liu@jerryjliu0XAI score41AIOpenDocRouter is a unified API and billing interface for document OCR models, ranging from lightweight open-source options like MinerU to frontier VLMs like Opus 5.5. Per the linked post, models are served at cost with a small transaction cut, rate limits are handled, and bounding boxes and layout are offered as a service.
Alexander Doria@DorialexanderXAI score46AILightOn has released LightOnOCR-3, a family of OCR models in 0.8B and 4B versions that it says lead benchmarks including OlmOCR-Bench and ParseBench, with the 0.8B model positioned as the sub-1B option. The models recognize text, handwriting, images, charts and document structure in one pass, process documents up to twice as fast as LightOnOCR-2, and are released under the Apache 2.0 license.
Dhravya Shah@DhravyaShahXAI score42AISupermemory introduces MemoryRepo.dev, an open-source implementation of Cognition's dreaming memory system built on Cloudflare Artifacts, Durable Objects, Alchemy, and Effect. The project follows Cognition's Devin memory design, which builds a memory graph across sessions and prunes stale records overnight. Supermemory says it will incorporate learnings from this research into its own product.
Zhihao Jia@JiaZhihaoXAI score62AILithos AI says it is open-sourcing lithos-metal, which uses megakernels and DSpark speculative decoding. The post claims Qwen3.8-27B reaches a peak of over 200 tokens per second per user on a single Apple M5 Max. It says users can try the tool with any coding agent in one command, and links to the code on GitHub and a technical blog.
Muse@MuseOfficialAI score20AIMuse, created by @JJEnglert, is trained on the user's movie and TV preferences and weekly curates a feed of films in theaters and on owned streaming services. Users can upvote or downvote picks to refine the feed, and when they want to see a movie, Muse books the ticket through a wallet link integration.
Rahul@sairahul1XAI score25AIA solo founder used Teamily AI, a platform where humans and AI agents share one group chat, to automate a multi-step project. A research agent analyzed the author's 20 most-saved posts, a writer agent drafted content, a web agent built a live website in Website Builder, and a custom Verification Editor agent blocked the launch twice over a misquoted Anthropic doc. The team can be saved as a Loop that reruns weekly and waits for human approval before publishing.
Rahul Bhardwaj@rahulb_coXAI score34AIListing Studio, an AI video tool for realtors, has launched in beta and is said to have already drawn over 700k views. The post says it lets agents create home listing videos without a film crew or editing.
Luke Edwards@lukeed05XAI score38AIPocketty launches as an SSH app for iPhone and iPad, built for herdr, that notifies users when an agent on any host is blocked. Tapping a notification opens the exact pane, and the app supports Tailscale and Bonjour natively, shows diffs for every agent turn, and requires no account or subscription.
Generalist@GeneralistAIOfficialAI score28AIGeneralist has announced GEN-1.5, its latest foundation model for the physical world. The post provides only a link to the company's blog for further details, so no specifications, benchmarks, or availability information can be confirmed from this source.
ATOMS@atoms_resXAI score25AICatalyst has raised $30M in funding backed by Sequoia Capital and other investors, according to the post. The company describes itself as an AI platform for financial research and automated trading, and the post links to a waitlist.
AIJevman is an open-source Pac-Man benchmark where AI models play 100 games each against the classic scripted ghosts. Each model gets a maze state at every junction and returns a direction probability, with answers over 2 seconds replaced by a backup rule. Community models can join the leaderboard by submitting games that CI replays to verify their scores.
Daniel Han@danielhanchenXAI score38AIUnsloth now supports OS-level sandboxing on Linux via bwrap, on Mac via seatbelt, and on Windows via Microsoft's MXC. Per-tool-call latency is under 100ms across all three, and its software-style sandboxing with regex AST checks adds about 3ms. The Windows integration was built in collaboration with Microsoft.
OpenClaw🦞@openclawOfficialAI score34AIOpenClaw released its v2026.9.9 patch, adding support for GPT-6.1 Sol in Codex and Claude Haiku 5.5. The update also improves recovery from failed updates, fixes missing iMessage replies, and resolves scheduled-job issues. The release credits 90 contributors.
elvis@omarsar0XAI score34AIVoice Arena's Monsoon ASR dataset fine-tuned Whisper Medium on Bengali FLEURS, reducing LLM word error rate from 85.27% to 7.65%. The corpus spans 100,000 hours across 50 languages, and Voice Arena says more than 80 organisations have asked to license it since its launch a week ago.
Unsloth AI@UnslothAIOfficialAI score44AIUnsloth now supports OS-level sandboxing on Windows by integrating Microsoft's open-source mxc repository for sandboxed code execution. The integration adds under 100 ms of overhead, according to the post. A setup guide is available in Unsloth's documentation.
AIHelm.ai said it signed $70 million in commercial contracts for its foundation models for physical AI over 12 months, spanning global automotive OEMs, Tier 1 suppliers, and industrial automation companies. The Redwood City, Calif.-based company said it has projects bound for production in autonomous vehicles, mining, and construction, and that it is on a path to break even. CEO Vladislav Voroninski said its models are trained on unsupervised "deep teaching" and are environment-agnostic.
AIWillow Care Inc., maker of the AI dictation app Willow Voice, chose CoreWeave for its forward-deployed support rather than compute alone, according to co-founder and CTO Lawrence Liu. Liu said CoreWeave's reinforcement learning infrastructure lets Willow focus on eval alignment, while Willow fine-tunes its own speech recognition model and pairs it with a compact post-processing LLM. He said inference demand is growing faster than training as dictation use climbs.
elvis@omarsar0XAI score32AIFish Audio's Drama 3 voice model lets users direct tone, emotion, and pacing in plain language, and can shift emotion mid-sentence. The poster, who found it remarkably effective in testing, says the control over delivery is unlike anything previously seen. A preview is available through the API as drama-3-preview.
elvis@omarsar0XAI score33AICogent Attack Path Analysis uses Cogent's own cyber models to find the longer attack routes an AI agent swarm could take into a company. It then identifies the single fix that closes each route, since agent swarms persist where human attackers give up.
boris@boristaneXAI score33AIPolylane has launched Polylane for Vercel, agents that monitor a user's Vercel account around the clock, detect issues, and fix them automatically. The launch is also running on Product Hunt, where the team is seeking a #1 ranking.
Santiago@svpinoXAI score22AIA security platform uses agents to map a system's potential vulnerabilities and identify routes an attacker could take to reach sensitive data. It then recommends changes to close those paths. The quoted post cites a 700-agent swarm that breached Hugging Face with over 17,000 actions, and presents this tool, Cogent Attack Path Analysis, as the defensive counterpart.
OpenCode@opencodeOfficialAI score32AIOpenCode is offering free access to its Step 5 Preview for one week. The announcement highlights a 1M context window, multimodal support, and zero data retention.
AIAtomic Canyon's Nuclear Industry Virtual Assistant (NIVA), developed with nuclear-industry groups, is now available to the entire U.S. fleet of 94 reactors after pilot testing at Constellation Energy plants. Nuclearn says its products have reached more than 65 U.S. partners, and the article says the industry is turning to AI to help manage regulatory paperwork and a shrinking, aging workforce.
AITessl's Code Review defines review standards as skills in the repository, called lenses, routed to files by a repo-owned profile file. Because these team-visible standards are portable, lessons from review can feed back into code generation and maintenance, not only the next review.
AIA developer reverse-engineered the firmware of a Nokia 110 4G and built a native AI chat app that runs on the phone. The app sends typed messages to the DeepSeek chat API and uses tool calls to check battery level, toggle the torch, start calls, and set alarms. It runs as a prototype loaded into RAM from a computer, so it must be reloaded after a restart or power-off.
prathosh A P@prathoshapXAI score34AILatentForce has launched Workspace, a tool where teams discuss work and the workspace records the decisions that agents then build from. The post argues that code is now cheap while attention and coordination remain scarce, and that agents currently build from outdated plans.
merve@mervenoyannXAI score37AIllama.cpp v0.6.0 adds support for multimodal Jev-like models and a performance upgrade for Metal, according to Merve Noyan. The post says models can be served with a single command, `llama serve -hf ggml-org/Clef-Flash-GGUF`, and points to a trending list of decision models on Hugging Face.
AIZhengxing Innovation launched a Physical AI solution at APRCE 2026 for retail human-robot collaboration, built on its "embodied brain" and comprising the H1 humanoid and C1 wheeled-arm robots plus the M1 management platform. The company says the solution needs no store renovation, reports 99% autonomous task completion, and plans commercial service in 2027 via direct purchase or RaaS subscription.
HeyGen@HeyGenXAI score22AIReal estate figure Ryan Serhant is launching a daily video series on sales, business, and personal branding, delivered by his official AI avatar rather than filmed by him. The launch is announced in a post linking to a Variety report, and the avatar is produced with HeyGen.
AI Edge@aiedge_XAI score34AIY Combinator CEO Garry Tan has open-sourced his complete Claude Code configuration as the gstack repository on GitHub. The post calls it one of the best GitHub repos ever, though it provides no details on the setup's contents or features.
X.PIN@thexpinXAI score40AITian Keyu's unnamed startup has raised nearly $30M from 5Y Capital and IDG at a $200M post-money valuation, according to Bloomberg. The NeurIPS 2024 award-winning researcher's 10-person team is developing a 200,000-symbol visual vocabulary to help AI process video. Tian claims the approach could cut video-generation costs at least tenfold, with a full model release planned for 2027 and no product yet.
vLLM@vllm_projectOfficialPickAI score62AIvLLM v0.31.0 is released with 717 commits from 307 contributors, including 96 first-time contributors. Highlights include DeepSeek-V4.1-Flash support, a vllm preload command that keeps weights in GPU memory across restarts, and Model Runner V2 with draft-model speculative decoding. The release also adds large-scale serving, scheduling, and HiSparse fixes, with full notes linked on GitHub.
Why it matters: The release lists concrete changes across serving, scheduling, and model support, which helps operators judge whether the upgrade affects their deployment path.
AIArchitect Financial Technologies has launched Liquid Inference, an LLM router that auctions each request to providers quoting the requested model, and the lowest qualifying offer wins. Buyers can set per-job cost caps, time-to-first-token limits, minimum throughput, and region or zero-data-retention rules, and the max price is locked before generation. The source states that fees, provider list, and latency data are not yet public.
AIHu Han, former head of Tencent Hunyuan's visual large model algorithm center, is raising tens of millions of dollars for a multimodal startup at a target valuation of several hundred million dollars, with Yuanshi Capital as financial advisor. Investors say Hu has spoken with several firms over the past month and has suggested his model could eventually be sold to large technology companies such as DeepSeek.
dex@dexhorthyXAI score22AIDeveloper Dex Horthy highlights Fold, a Effect-based agent harness by @0xblacklight that stores state as a portable append-only log. It runs on Cloudflare Durable Objects out of the box and includes a retro OpenTUI interface with 10 themes.
AIMastra launched the Mastra Agency Program, a network of certified agencies and consultancies that build Mastra agents for clients. The launch includes five partners: Deerfield Group, Blue Drop Labs, Frontleap, Handpicked, and Young Security. Every partner has been vetted by Mastra's FDE team and receives direct access to Mastra's leadership and regular roadmap updates.
AIStep Terminal will unveil its first large-model-native agent smartphone, the STEPX Neo, at a "Ready Builder One" launch event in Shanghai on October 13. The company says the device is built agent-native across its model, system and hardware, and the event will also announce the latest progress in its ecosystem partnerships.
AIUnsloth Studio binds remote-code approval to a fingerprint of the scanned code, so changed code requires fresh consent before it runs. It also blocks weight files that Hugging Face has flagged for malware in the path the selected loader would deserialize. The article describes these checks as one layer among several, alongside package-content scans and OS sandboxes, and notes that the scanner is not a sandbox and cannot catch every evasion.
AIMUZIM L1 is a desktop data dock with up to 24TB of storage, dual SSD slots, and a Vibe Search feature that finds files by natural-language description, with local-first processing rather than default cloud upload. Lumeria Lumoscope is a multispectral skin scope that clips onto a phone, using RGB, ultraviolet, polarized, and near-infrared light, priced at $199 in pre-sale. The article also covers immurok IK-1, a 59-dollar wireless fingerprint key with a 60-day standby battery that authorizes sudo, SSH, and Git actions on Mac, Windows, and Linux.
Inferact@inferactOfficialAI score38AIInferact, working with DeepSeek, NVIDIA, and SemiAnalysis alongside the vLLM community, says joint work across models, custom kernels, and engine serving cuts time to first token (TTFT) by nearly 70% at ~100K throughput. vLLM is the open-source inference engine, and Inferact optimizes it for enterprise production deployments.