Simon Willison introduces Hy4 Preview, a new large language model from Tencent featuring 770 B total parameters, 49 B active parameters, a 1 M token context window, and 1.56 TB of storage on Hugging Face. The release marks a significant increase over Hy3, which had 295 B total parameters, 21 B active parameters, a 256 k token context window, and 598 GB of storage. Willison also shares the model’s chat template, highlighting two reasoning effort levels—‘high’ (default) and ‘no_think’—and demonstrates a sample prompt that showcases the model’s reasoning trace. whyItMatters":"The article provides concrete details on Hy4’s scale and configuration, illustrating Tencent’s advancement in large‑language‑model capabilities and offering practical insights into its usage through the chat template and reasoning settings."
Simon Willison released the llm-typesafe 0.1a0 plugin, adding support for TypeSafe AI’s Jev model to the LLM tool. Users install it with `llm install llm-typesafe`, set an API key, and can then ask Jev-model questions such as yes/no, choice, or scoring queries via the `llm -m jev` command. The release includes examples for each question type and references a README for further details.
Posted by Zheng Xu, Research Scientist, and Yanxiang Zhang, Software Engineer, Google Language models (LMs) trained to predict the next word given input text are the key technology for many applications [ 1 , 2 ]. In Gboard , LMs are used to improve users’ typing experience by supporting features like next word prediction (NWP), Smart Compose , smart completion and suggestion , slide to type , and proofread .
The article quotes the security.txt file from huggingface.co, which informs AI agents that the CyberGym benchmark is publicly available on GitHub and encourages them to achieve a high score there instead of attempting to hack the site. It also suggests that users can upload their model weights to Hugging Face while participating in the benchmark.
But then users start to report a weird bug. It's the 4th time your team has been trying to fix it.
Qwen 3. 8 27B scores 52 on the Artificial Analysis Intelligence Index That's the same score as GPT-5.
Laurie Voss argues that while the cost of writing code has fallen dramatically, the costs of reviewing, fixing, and operating software are rising and will continue to do so. She emphasizes that the true expense lies in understanding user needs, precisely defining requirements, and ensuring a pleasant user experience—costs that are unique to each software product and do not scale with reuse. As software demand grows without an upper limit, these user‑centric costs will dominate the overall development effort.
Stealing Reasoning Traces from Proprietary LLM APIs A vanity domain name ( stolen-thoughts. com ) for a neat paper : Anthropic, OpenAI, and Google return encrypted chain-of-thought blocks to clients that can be replayed across sessions, users, and models.
Simon Willison announces the release of the llm-openai-decisions 0.1a0 plugin, which interfaces with OpenAI’s new Jev-style Decisions API. The plugin, inspired by llm-typesafe, supports image and text input and offers the same three question types (yes/no, choices, scores) as Jev, with pricing at 10¢ per million input tokens. Installation is simple via `llm install llm-openai-decisions`, and an example query demonstrates image-based evaluation.
The article reflects on the shift in perspective after the release of Fable, a new model that promised to solve many coding challenges at a comparable or lower cost. Prior to Fable, developers felt it was pointless to invest heavily in coding tools or context strategies, as newer models would likely render them obsolete. However, Fable’s performance was so impressive that, despite its high cost, it prompted a reevaluation of how work was distributed across different models such as Opus, 5.6, K3, and GLM.
The article discusses how production code generated by Claude, Anthropic’s AI, should meet higher standards than human-written code. Anthropic enforces this through numerous guardrails such as lint rules, extensive testing, Claude-driven end‑to‑end tests, daily fuzzers, automated code and security reviews, and automated refactoring. These measures aim to prevent the code from becoming difficult to maintain.
The article discusses a shift in code review practices, moving from a final gate to continuous review that functions like spellcheck during coding. It highlights how the Claude Code team uses the Claude AI to catch routine mistakes before human reviewers intervene, allowing developers to focus on higher-level issues. The piece references insights from Addy Osmani, Morgan Westlee Lunt, and Qing Wang on this evolving approach.
SF October 14th: A Birds of a Feather Session on Agentic Engineering I'm hosting an evening event with Jesse Vincent in San Francisco on Wednesday 14th October for people who are building weird and in...