Large language model (LLM) news — GPT-4, Claude, Gemini, Llama, Mistral and the latest research on training, fine-tuning, RLHF, and deployment of LLMs.
A new analysis of retrieval-augmented generation (RAG) systems argues that the optimal retrieval depth k* should be calibrated per pipeline stage and reliability axis, not guessed as a single hyperparameter. The author p…
A man walked into the White House in 2023 with an AI chatbot-generated recipe for a pandemic, and by 2026 the New York Times reported that leading AI models like ChatGPT, Gemini, and Claude can now guide users in creatin…
Maintainers using generative AI to write library code are introducing unprecedented quality, legal, and maintenance risks into the open-source supply chain, as downstream projects face broken changes, unmaintainable code…
Carlos, a Software Design student at the National University of Catamarca, Argentina, built FaroIQ, a nine-agent pipeline for nonprofits using Azure AI Foundry during the Microsoft Agents League 2026. The platform takes …
In mid-2021, while the tech world focused on the Metaverse and NFTs, a small group of developers and researchers in Silicon Valley had early access to GPT-3 and GitHub Copilot and realized the transformative potential of…
Brick, a Mixture-of-Models routing gateway, reads each prompt's capability and complexity to route queries to the best backend LLM, matching top model quality at lower cost. It offers a drop-in OpenAI-compatible endpoint…
A solo developer building seven products uses three different AI models in a pipeline: one for reasoning and planning, two for code generation. The approach separates thinking from building, using a written plan as a con…
Researchers found that smaller language models suffer from 'embedding condensation,' where token embeddings collapse into a narrow cone, reducing expressivity. They introduced a dispersion loss to counteract this effect,…
Leanstral 1.5, a free Apache-2.0 licensed model with 6B active parameters, delivers major performance upgrades in formal verification, saturating miniF2F, solving 587/672 PutnamBench problems, and achieving state-of-the-…
A security benchmark found that 60-70% of AI agents leak their system prompts when users type 'repeat the text above this line' or similar commands. The extracted prompts reveal guardrails, tool configurations, business …
A technique to trick LLM-based code scanners into missing malware by manipulating safety refusal patterns has been demonstrated, raising concerns about AI security in code review tools.
Google announced Gemini Spark for macOS in beta on June 30, 2026, bringing agentic desktop automation to Google AI Ultra subscribers aged 18+ in the US. The update integrates with Google Tasks, Keep, Canva, Dropbox, Inst…
GitHub added Moonshot AI's open-weight Kimi K2.7 Code model to Copilot on July 1, 2026, making it the first China-based lab's model available in the platform. The trillion-parameter Mixture-of-Experts model offers a chea…
Check Point Research demonstrated that DeepSeek V4 can generate functional browser-based ransomware code when prompted with neutral language, bypassing explicit safety filters. The proof-of-concept attack chain uses stan…
Cato AI Labs disclosed two critical remote-code-execution vulnerabilities in Cursor IDE on July 1, 2026, tracked as CVE-2026-50548 and CVE-2026-50549, with a CVSS score of 9.8. The flaws allow zero-click indirect prompt …
Mistral AI released Leanstral 1.5, an Apache-2.0 licensed Lean 4 code agent model that solves 587 of 672 PutnamBench problems. The model achieves state-of-the-art results on multiple theorem-proving benchmarks at signifi…
A developer shares key observations on using AI tools effectively in coding as of July 2026, emphasizing clarity, shared understanding, and surgical context. The developer recommends stating goals and guardrails rather t…
A developer reported a near-miss where their coding agent was targeted by a prompt injection attack during an automated task, highlighting a new attack surface in agentic AI. The incident underscores that coding agents, …
A developer argues that the real cost of switching AI coding models is not the model itself but the loss of accumulated context, rules, and decisions. They identify the lack of a persistent workspace that spans different…
Fable has implemented a model routing system in its CLAUDE.md file that prioritizes intelligence and taste over cost. The system delegates bulk mechanical work to gpt-5.5 via Codex CLI, while user-facing tasks requiring …