tools
DeepMind’s chemistry agent proposes molecules in an hour, with laboratory validation still ahead News
DeepMind describes a roughly one-hour computational chemistry workflow that generates testable molecule proposals rather than experimentally validated drugs.
Vercel Labs’ Agent Desktop reaches 1.0 with structured desktop control News
Vercel Labs released Agent Desktop 1.0, a Rust command-line tool and C library that let agents inspect accessibility trees and operate desktop controls.
Dream Labs pairs a $90 million seed round with a cross-robot action API News
Dream Labs has announced a $90 million seed round and a public Dream-1 API that uses shared end-effector goals to attempt control across different robots.
An AI-built Enron wiki tests whether citations actually support its claims News
An Enron email wiki reports improved source support after a second verification pass, while its audit still finds unsupported details in cited facts.
OpenAI makes its Decisions API generally available for bounded workflow choices News
OpenAI’s Decisions API is generally available for classification, routing and other bounded choices, with a company-claimed speed advantage over Responses.
Mistral’s Voxtral 3 packages voice agents, with latency and openness caveats News
Mistral launched Voxtral 3 and a voice-agent SDK that connects listening, reasoning, tool calls and speech, while public weight access remains unverified.
DeepSeek opens Search and Fetch at $0.75 per thousand successful calls News
DeepSeek opened separately billed Search and Fetch tools at $0.75 per 1,000 successful requests per endpoint, excluding errors, empty searches, and timeouts.
Google Playground combines prompt-built games with a publishing network News
Google launched Playground for AI-assisted browser-game creation, adding exportable game files, visibility controls and a social discovery layer.
OpenAI launches a Decisions API that returns choices instead of prose News
OpenAI’s Decisions API beta returns bounded decisions from text and images at $0.10 per million input tokens, with no output-token charge.
Ai2 releases AstaBrief for faster cited scientific reports News
Ai2 has released AstaBrief, an open-weights report writer that turns a question and retrieved scientific excerpts into a cited report in one pass.
Comfy launches an agent that builds and repairs visual workflows News
Comfy Agent ships in Comfy Cloud to build, edit, repair, and run workflows, with Desktop support promised later.
llama.cpp adds GLM-5.3-Flash support, bringing a 328 GB model into the local-runtime ecosystem News
llama.cpp merged support for Z.ai’s MIT-licensed GLM-5.3-Flash on September 30, but the official model repository still occupies 328 GB.
Stripe agrees to acquire OpenRouter as it reports 10 trillion tokens a day News
Stripe has agreed to acquire OpenRouter, whose official announcement says its routing platform processes more than 10 trillion AI tokens daily across 400-plus models.
LabFactory tests whether an AI can build a runnable scientific solver News
LabFactory asks an AI builder to package code, models, retrieval and tools into a scientific solver that a separate host executes on held-out inputs, shifting evaluation from persuasive answers to working artifacts.
Qwen releases Image-2.1 weights, but commercial use needs a separate licence News
Qwen-Image-2.1 ships a 33.1 GB local image-generation and editing pipeline with native transparency, but its research-only licence bars commercial use without separate permission.
Bonsai 2 ships a 27B local model in a 5.95 GB download News
Prism's ternary Bonsai 2 compresses its smallest 27B language-weight pack to 5.95 GB, but the headline is a disk-size claim rather than a full runtime or VRAM requirement.
Retriever launches free AI tasks funded by sponsored cards beside results News
Retriever says its Free Mode runs everyday AI tasks at zero credits with fair-use limits and a clearly labeled sponsored card displayed beside the result.
Anthropic shipped a content checker that cannot tell you if Claude wrote it News
Anthropic launched a free browser-based tool that reads content credentials embedded in files, and the page states plainly that it cannot determine whether Claude was involved in creating the content it checks.
The Codex desktop app ships a full copy of LibreOffice News
Simon Willison found on September 1, 2026 that OpenAI's Codex desktop app caches 1.7 GB of bundled runtimes -- a complete Python installation, a complete Node.js installation, and native binaries for git, Poppler and the entire LibreOffice office suite -- which the agent's skills then invoke to handle documents.
A 104 GB model now runs on a 48 GB Mac by streaming experts off the SSD News
slotstream, a single Swift binary released as a Show HN on September 1, 2026, runs the 104 GB Qwen3.8-Flash-Next mixture-of-experts model on Macs with a fraction of that memory by keeping a small trunk resident and reading expert weights off the SSD on demand -- about 12 tokens per second on a 48 GB machine.
Two-bit models now run on every major llama.cpp backend News
The official 2-bit weight format merged into llama.cpp now works on mainline CPU, Metal, Vulkan, and CUDA after the CUDA path landed upstream, completing a backend sweep that started as an ARM-only contribution.
AskChem indexes 2.4 million individual chemistry claims instead of 147,000 papers News
A team at NYU built a live chemistry search system that breaks each paper into atomic claims, each carrying its own source identifier and verbatim quote, so a researcher or an AI agent retrieves the individual assertion rather than a ranked list of documents.
A 26-billion-parameter model runs in 2GB of RAM by streaming experts off the SSD News
TurboFieldfare, an open-source Swift and Metal runtime, runs Gemma 4's 26-billion-parameter model on an 8GB MacBook Air by keeping only a 1.35GB core in memory and pulling each token's experts from disk as it needs them.
A Show HN Promised Frontier Quality for Half the Cost. Its Repo Describes a Router. News
World Model Optimizer turns agent traces into a policy that sends easy requests to cheap models and hard ones to a frontier endpoint, but its headline savings claim shifted between the repository and the post, and no distilled model has been released.
Capital One Open-Sources VulnHunter, an AI Agent That Hunts Security Bugs News
Capital One released VulnHunter, an open-source agentic AI security tool that reasons like an attacker, tries to disprove its own findings before reporting them, and has been run across thousands of the bank's own repositories.
Biology becomes AI's next benchmark battleground -- and today's agents are failing News
New benchmarks show frontier AI agents scoring as low as 17% at basic biology data retrieval and returning wildly different answers to the same query, but a single deterministic lookup tool pushes accuracy above 90% -- as OpenAI launches GeneBench-Pro to measure judgment-heavy biology.
Anthropic launches Claude Science, an AI workbench built for biologists News
Anthropic released Claude Science, a customizable research workbench that wires Claude into more than 60 scientific databases and tools like PubMed, Jupyter, and R, and is giving qualifying researchers up to $2,000 in compute.
Anthropic's Claude Science puts a whole lab bench inside the AI News
A new workbench pulls a scientist's scattered tools - literature, notebooks, cluster jobs - into one place and keeps a full, checkable record of how every result was made.
A model that rivals the frontier now squeezes onto a single high-end desktop News
Aggressive compression shrinks GLM 5.2 by more than 80 percent while keeping most of its accuracy, putting a near-frontier model within reach of local hardware.
An open-source 'AI crew' that turns a coding assistant into a video studio News
A project called OpenMontage shot to the top of GitHub in a day, claiming to be the first open-source system that lets AI agents handle a whole video production from script to final cut.