Anthropic today announced it's releasing Claude Sonnet 5.5, a faster and more efficient update to its mid-tier, workhorse AI model.
Anthropic says Sonnet 5.5 generates output more than 30% faster than Claude Sonnet 5 and can reduce the total cost of completing a task by as much as 30%, primarily because it uses fewer tokens and fewer tool calls rather than because of a lower API sticker price.
The Nvidia Open Agent Safety Platform combines OpenShell 0.1.0, the Apache 2.0 agent runtime the company announced at GTC in March, with Nvidia Sentry, a watchdog service that runs on the company’s BlueField-4 data processing units (DPUs).
The new OpenShell release adds a policy prover that checks that an agent’s various permissions can’t be combined into something the operator didn’t intend — like hacking HuggingFace.
The AI Automators explores the implementation of Jev, a steerable decision model designed to enhance agentic Retrieval-Augmented Generation systems. This approach replaces traditional large language model processes with high-speed, cost-effective micro-decisions for tasks like document re-ranking, citation verification, and model routing to optimize overall system performance and efficiency.
Sam Witteveen demonstrates building a local model router that intelligently routes tasks to either local or cloud-based LLMs. The system categorizes requests, evaluates difficulty, and detects private information to optimize performance, token usage, and data security while utilizing Jev or open-source alternatives as a decision-making layer.
OpenAI releases GPT-6 Sol and Luna models
OpenAI is expanding its lineup of AI models with the new GPT-6 Sol and GPT-6 Luna, mid-priced and lower tier models designed to serve as enterprise workhorses for common tasks requiring less intelligence — and both are priced at half or less the costs of their predecessors when accessed over OpenAI's application programming interface (API).
Anthropic releases Opus 5.5 with lower prices
Anthropic’s newest model, Opus 5.5, was released on Tuesday, setting a new state-of-the-art in coding and knowledge work performance, according to the company. Opus is the most capable and expensive tier in Anthropic’s three-tier Claude lineup; Sonnet sits in the middle, while Haiku is the fastest and cheapest. Notably, Anthropic says, the release outpaces the larger Fable model in many benchmarks and succeeded in a number of informal tasks that Fable failed to complete.
Xiaomi has released and open-sourced MiMo-V2.6 Pro and Flash, two omnimodal models now available in AI Studio, MiMo apps, its API and OpenRouter.
Sam Witteveen explores Jev, a new system one model designed for fast, non-autoregressive classification tasks in software. By moving away from traditional chat interfaces and long-form text generation, this approach offers a function-call-like method for decision-making that prioritizes speed and efficiency over complex reasoning chains.
On Thursday, PrismML released Bonsai 2 27B, its latest in a family of models, which compresses Qwen3.8 27B, a widely used open source model from Alibaba, down to 5.9 GB. That’s small enough to fit on a PC and, possibly, a high-end smartphone. It’s a 9x to 10x reduction in memory versus the original.
Canadian AI startup Cohere says it can do enterprises one better, assuring them their data remains private and off-limits to itself and cloud providers using Model Vault, the single-tenant inference product it already sells, and a technique called confidential computing.
The AI industry is experiencing a fundamental shift from model training to AI inference, driving unprecedented hardware innovation as companies race to optimize for real-time, large-scale LLM applications.
Kai examines the technical requirements and hidden infrastructure demands of running DeepSeek V4.1 Flash on personal hardware. The analysis covers the mixture-of-experts architecture, memory allocation challenges, and the practical trade-offs between local hosting versus API usage for developers.
Sam Witteveen explores the MiniCPM 2.5B model, examining how scaling its data and training recipe from the previous 1B version affects its performance in function calling and agentic tasks. The analysis includes a look at new RL2 training techniques, benchmark results, and practical testing for sub-agent applications using RunPod serverless deployments.
DeepSeek launched DeepSeek-V4.1-Flash last night with a 552-billion-parameter mixture-of-experts backbone, native vision, a 1-million-token context window and an architecture built to make repeatedly reading large contexts cheaper.
Its open weights are are available for developers and enterprises to download and use for commercial purposes under a permissive, enterprise-friendly MIT License on Hugging Face. For developers evaluating the model for coding agents and other long-running workflows, however, the headline API rate is immediately enticing.
Kai examines the technical shift enabling large-scale artificial intelligence models to run on consumer-grade hardware. By analyzing new architectural developments in model storage and memory tiering, this exploration breaks down the trade-offs between local inference costs and processing speeds when handling complex tasks compared to traditional cloud-based solutions.
AI LABS explores how Graft optimizes coding agents by building a project knowledge map, reducing token usage, and increasing execution speed. This open-source tool allows agents to locate specific code dependencies more efficiently, addressing the context window issues often encountered during complex development tasks.
Sam Witteveen explores the release of PAIR, an open-source tool designed to manage local AI agents across multiple household devices. The video explains how this virtual inference router enables parallel task delegation, distributing workloads across various hardware configurations to improve efficiency for agentic software workflows.
The rumors were true, all of them (and then some): OpenAI today is releasing GPT-6 Astra, a new frontier model that the company says likely marks the onset of artificial generalized intelligence (AGI), its long sought goal of "highly autonomous systems that outperform humans at most economically valuable work."
In a closed a press briefing earlier today, OpenAI co-founder and president Greg Brockman offered an unusually direct formulation of that message, ending the session with: “Welcome to the AGI era.”
Google hasn’t released a frontier-level Gemini Pro AI model since early 2026, but it sure loves rolling out new Gemini Flash variants. Today, Google is announcing its third Flash model release in just six weeks, making it more likely that we’ll never see the promised Gemini 3.5 Pro. But no matter, says Google, because Gemini 3.8 Flash is its best reasoning and coding model yet.
Gemini 3.8 Flash comes in two variations. There’s the standard Flash, which Google describes as a “workhorse” model that’s good for anything from agentic tasks to software development. Then we have Gemini 3.8 Flash Cyber, which runs on the same foundations but has been tuned for vulnerability detection and mitigation.
On Tuesday, Anthropic released Fable and Mythos 5.1, twinned versions of the company’s most advanced AI model. In addition to performance upgrades, the new Fable release includes changes meant to reduce token cost and false-positive restrictions from the model’s safeguards.
Kai examines why local AI coding agents on Mac devices often experience significant performance degradation over time due to inefficient prefix caching. The guide details how oMLX implements a two-tier caching system to address this, along with instructions for integrating the tool with popular coding environments to maintain consistent response speeds throughout long sessions.
