AI/TLDR — every new AI model, tool, repo & paper
The latest AI releases, refreshed every 2 hours and explained in plain English.
What AI shipped today?
In the last 24 hours AI/TLDR tracked 9 new AI releases, including GPT-5.6 — OpenAI previews Sol, Terra, and Luna tiers, 1littlecoder: 'GPT 5.6 — What, Availability, Pricing' and Ornith 1.0 — open-weight coding models that learn their own RL scaffold. AI/TLDR is an AI release tracker that follows new AI models, open-source tools, papers, datasets and benchmarks — refreshed every 2 hours from verified primary sources and explained in plain English.
AI Release Index — live stats on AI releases · Learn AI
- GPT-5.6 — OpenAI previews Sol, Terra, and Luna tiers
OpenAI announced GPT-5.6 with three named tiers — Sol (flagship), Terra (balanced), and Luna (cheap and fast) — adding new max and ultra reasoning modes. Access starts as a limited preview for trusted partners.
- 1littlecoder: 'GPT 5.6 — What, Availability, Pricing'
1littlecoder walks through OpenAI's same-day GPT-5.6 announcement — the Sol, Terra, and Luna tiers, the new max and ultra reasoning modes, the published per-million-token prices, and who actually gets access during the limited preview.
- Ornith 1.0 — open-weight coding models that learn their own RL scaffold
Ornith 1.0 is an MIT-licensed family of agentic coding LLMs (9B, 31B, 35B MoE, 397B MoE) whose RL loop writes its own task-specific scaffold instead of using a fixed human-designed harness.
- Sam Witteveen: 'Introducing Ornith 1.0' — open-weight coding LLM walkthrough
Sam Witteveen walks through Ornith 1.0, DeepReinforce's MIT-licensed coding model family whose RL loop learns its own scaffold — uploaded hours after the 9B–397B weights landed on Hugging Face.
- Un-0 — image generator built from coupled oscillators, not a neural net
Un-0 generates images by simulating a population of Kuramoto oscillators whose pairwise coupling strengths are the learned parameters. It hits FID 6.74 on ImageNet 64x64 with MIT weights and training code released.
- Lilian Weng: 'Scaling Laws, Carefully' — first new Lil'Log post in 13 months
Lilian Weng walks through scaling laws end-to-end: why Kaplan and Chinchilla reached opposite conclusions, how parameter counting and fit region change the answer, and what data scarcity means for the curves now.
- GitHub Copilot for Jira — GA brings agent streaming inside Atlassian
GitHub Copilot for Jira is generally available, letting teams watch the coding agent run on a Jira ticket and steer it in Jira chat without switching to GitHub.
- Wes Roth: 'OpenAI JUST announced JALAPENO'
Wes Roth breaks down OpenAI's new Jalapeño inference chip, built with Broadcom, and what a custom OpenAI accelerator means for the AI hardware race.
- OpenKnowledge v0.18.0 — local-first AI markdown wiki with codebase generator
OpenKnowledge v0.18.0 ships a codebase-wiki starter pack that scaffolds an agent-written wiki of a repo, plus AI editing with Claude, Codex, and Cursor in a local-first markdown app.
- Sam Witteveen: 'Qwen-AgentWorld The World Model for RL Environments'
Sam Witteveen's June 25, 2026 walkthrough of Qwen-AgentWorld — Alibaba's open-weight world model that simulates seven agent domains — landed hours after the paper trended on Hugging Face.
- Anthropic accuses Alibaba Qwen of largest-ever Claude distillation attack
Anthropic told U.S. senators that operators tied to Alibaba's Qwen lab ran 28.8 million Claude conversations across nearly 25,000 fake accounts between April 22 and June 5, 2026 — its largest known distillation attack on Claude.
- Qualcomm to Acquire Modular — $3.9B all-stock deal for Mojo and MAX AI stack
Qualcomm will buy Modular for about $3.9 billion in all stock, taking ownership of the Mojo language and the MAX AI platform built by Chris Lattner to run AI models across CPUs, GPUs, and NPUs.
- 1littlecoder: 'Unlimited OCR in 6 mins!'
1littlecoder walks through Baidu's new Unlimited-OCR — the 3B vision model that parses long documents in a single pass — in a 6-minute hands-on, hours after the model trended on Hugging Face.
- GitHub Copilot CLI GA — tabbed terminal with MCP, skills, and plugins
GitHub Copilot CLI's new terminal UI is generally available, adding Session/Issues/PRs/Gists tabs, in-session /mcp, /skills, /plugin, and /settings commands, and theme-aware accessibility with screen-reader support.
- GitHub Copilot app gets BYOK — Anthropic, Ollama, and LM Studio supported
The GitHub Copilot desktop app now supports bring-your-own-key for Azure OpenAI, Anthropic, Microsoft Foundry, Foundry Local, LM Studio, Ollama, and any OpenAI-compatible endpoint, with keys stored in the OS keychain.
- Mistral Connectors — admin controls, multi-account, and a new debugger
Mistral expands its Connectors stack with per-workspace admin controls (GA), API keys scoped to specific connectors (GA), multi-account auth (GA), an 11-step Connectors Debugger (preview), and connector support inside Vibe Code and Workflows.
- OpenAI Jalapeño — first custom inference chip, built with Broadcom
OpenAI Jalapeño is OpenAI's first custom inference chip, co-designed with Broadcom and built by Celestica. The ASIC targets LLM inference at substantially better performance per watt; engineering samples already run GPT-5.3-Codex-Spark.
- Gemini 3.5 Flash gets Computer Use — native browser, mobile, and desktop agents
Google DeepMind built Computer Use directly into Gemini 3.5 Flash, so the main Flash model can now drive a browser, an Android phone, or a desktop on its own through the Gemini API and Enterprise Agent Platform.
- Krea 2 — open-weight 12B image model with 2-second Turbo variant
Krea 2 is a 12B Diffusion Transformer text-to-image model released as open weights in two variants: Raw for fine-tuning and Turbo, which generates 2K images in about 2 seconds.
- OpenMontage — AGPL agentic video studio for AI coding assistants
OpenMontage is an open-source agentic video production system that turns Claude Code, Codex, or any AI coding assistant into a full video studio with 12 pipelines, 52 tools, and 500+ skills.
- Qwen-AgentWorld — language world models that simulate seven agent domains
Qwen-AgentWorld is a pair of open-weight world models (35B-A3B and 397B-A17B) that simulate seven agent environments — MCP, search, terminal, software engineering, Android, web, and OS — through chain-of-thought reasoning.
- OpenAI Daybreak — GPT-5.5-Cyber and Patch the Planet go live
OpenAI expands its Daybreak security program with the full release of GPT-5.5-Cyber, an updated Codex Security plugin, a partner program with CrowdStrike, Sophos, and Fortinet, and Patch the Planet, an open-source fix-funding effort with Trail of Bits.
- Nathan Lambert: GLM-5.2 — the step change for open agents
Nathan Lambert argues GLM-5.2 is the first open-weight model that feels right in coding harnesses as a general agent, matching closed leaders like Claude Opus 4.8 about seven months after they shipped, in what he calls a DeepSeek R1-style threshold moment.
- David Rosenthal: 'AI's Affordability Crisis' — the 70x subsidy that can't hold
David Rosenthal pulls together SemiAnalysis and Ed Zitron numbers to argue AI tokens are sold at a fraction of cost — Anthropic up to 40x, OpenAI up to 70x — and that real billing would turn a $200 ChatGPT plan into a $14,000 bill.
- Latent Space: 'Red-Teaming after Mythos' — Gray Swan on AI security
Latent Space hosts Zico Kolter (OpenAI board, CMU) and Matt Fredrikson (Gray Swan CEO) to argue AI security is not 'cybersecurity with AI' — Gray Swan's Shade red-teaming model now beats human attackers at breaking frontier LLMs.
- Claude Tag — Anthropic's @Claude Slack agent for shared teamwork
Claude Tag is a Slack agent from Anthropic that anyone in a channel summons by tagging @Claude. The shared bot breaks a request into stages, runs the work with tools an admin scoped per channel, and posts results back to the thread.
- Fireship: 'Midjourney wants to delete 30% of all death…'
Fireship reacts to Midjourney Medical's pitch — a 60-second full-body ultrasonic CT scan planned for a San Francisco spa — and the company's bold claim that AI-driven early diagnostics could prevent a large share of premature deaths.
- Simon Willison: 'Prompt Injection as Role Confusion'
Simon Willison highlights a new paper by Charles Ye, Jasmine Cui, and Dylan Hadfield-Menell arguing prompt injection is really 'role confusion' — language models lean on style cues, not content, to tell trusted text from user input.
- Armin Ronacher: 'The Coming Loop' — why even skeptics end up looping
Armin Ronacher argues 'harness loops' — outer systems that re-run AI agents past their natural stopping point — work well for code porting and benchmark runs, but breed defensive, dependency-creating code when pointed at real codebases.
- Mistral OCR 4 — 170-language document model with bounding boxes and confidence scores
Mistral OCR 4 extracts text plus per-block bounding boxes, type labels, and confidence scores across 170 languages, scoring 85.20 on OlmOCRBench and 93.07 on OmniDocBench at $4 per 1,000 pages.
- Wes Roth: 'Cursor JUST beat EVERYONE…'
Wes Roth's new video argues Cursor has pulled ahead of rival AI coding agents, walking through the Cursor Compile 26 opening keynote and the Composer 2.5 in-house coding model.
- Baidu Unlimited-OCR — 3B vision model parses long documents in one pass
Baidu's Unlimited-OCR is a 3B vision-language model that introduces Reference Sliding Window Attention to keep a constant KV cache, letting one forward pass transcribe dozens of document pages within a 32K context. Code and weights ship under MIT.
- PP-OCRv6 — PaddlePaddle ships 50-language OCR family from 1.5M to 34.5M params
PP-OCRv6 is the next PaddleOCR family with Tiny (1.5M), Small (7.7M), and Medium (34.5M) tiers covering 50 languages. The Medium tier lifts detection Hmean to 86.2% and recognition accuracy to 83.2%, gains of 4.6 and 5.1 points over PP-OCRv5_server.
- Oak — version control built for AI coding agents
Oak is a new version control system for AI coding agents that mounts repos lazily, runs branch-per-task, and benchmarks up to 95% faster than Git on snapshots, large binaries, and dirty trees.
- Anthropic-Cybersecurity-Skills v1.3.0 — 817 security skills across 6 frameworks
Mahipal Jangra's open Anthropic-Cybersecurity-Skills library jumps from 762 to 817 agent skills in v1.3.0, adding AI Security, Supply Chain, and Hardware/Firmware domains plus MITRE F3 as a sixth framework mapping.
- Simon Willison — porting Moebius image inpainting to the browser via Claude Code
Simon Willison shows how he used Claude Opus 4.8 to port the 0.22B Moebius image-inpainting model from PyTorch/CUDA to a browser-only WebGPU + ONNX demo, with the agent doing the framework conversion, weight upload, and UI work.
- Claude Code 2.1.186 — MCP login CLI plus auto-reply to bash commands
Claude Code v2.1.186 adds claude mcp login/logout for CLI-based MCP server auth, makes ! bash commands auto-prompt Claude to respond, and fixes 20+ background-agent and post-sleep streaming bugs.
- OpenAI Codex — SSD-burning SQLite log bug patched after 640 TB/year reports
OpenAI Codex CLI shipped two patches that cut about 85% of its SQLite log writes. Users had measured 37 TB written in 21 days, on track for 640 TB a year and full-drive SSD wear in months.
- Two Minute Papers: 'DeepSeek Just Solved AI's Billion Dollar Problem'
Two Minute Papers walks through the DualPath paper, which attacks the KV-cache I/O bottleneck behind agentic LLM serving costs and reports up to 1.96x higher online throughput.
- Sakana Fugu — multi-agent orchestration model that matches Fable 5 on quality
Sakana AI launched Fugu and Fugu Ultra, a multi-agent orchestration model delivered as one OpenAI-compatible API. Fugu Ultra coordinates a pool of expert agents and is reported to match Fable 5 on coding, reasoning, science, and agentic benchmarks.