Category
Modal hit $1B ARR and a $16B valuation, tripling both usage and valuation in five months. The growth came from sandboxes for AI-written code. Investors appear to be pricing execution infrastructure as a defensible layer of the agent stack, separate from intelligence.
Simon Willison asked ChatGPT to poll a GitHub repo hourly until Python 3.15 appeared. The small request shows where agents earn their keep: watching the lag between an upstream release and the moment your own tools can use it.
Anthropic priced Claude Haiku 5.5 to match OpenAI's GPT-6 Luna, and early coverage calls it better. That moves the competitive line in the tier where most agent labor actually runs.
OpenAI priced GPT-6.1 Sol at $2/$10 per million tokens against Astra's $10/$50, while claiming wins on agentic benchmarks. The price is the story: a volume-first bet on agent workloads that should change how you budget reasoning-heavy tasks.
Claude Cowork and chat are now one Claude, and the app keeps working after you close your laptop. The convenience story is real. The story nobody is telling is that the session boundary was doing security work, and it is gone.
OpenAI's biggest launch ever isn't a story about model specs. It's the moment the industry admitted the model layer no longer bottlenecks what your agent can do. The constraint moved to deployment, policy, and trust.
Greg Brockman is the face of OpenAI's Astra launch. The strategic tell isn't the model's raw capability. It's the growing case for alignment and controllability as the layer where value now accrues.
The new --restricted flag lets you strip an agent's ability to run commands or fetch the web before it starts. It's a small feature that flips the agent trust model from top-down to user-controlled.
OpenAI's Agents SDK now ships testing tools that let you validate agent workflows without a live model, sandbox, or network connection. That's not a convenience feature. It's an admission about who owns the layer that matters.
Google's Gemini 3.7 Flash reclaims ground it lost to Claude and GPT, reopening a three-way race for the model that powers the next generation of consumer agents.
Grok @Bot's launch reads like another model release. It's actually the first sign that the multiagent coding category has converged on a handful of vendors who can ship.
The consensus reads ChatGPT Work as a productivity update. The structural story: OpenAI is moving the agent from a tool you summon to the default place work gets delegated.
Qwen 3.8 Max is a 2.4T open-weight model competitive with closed frontier models. For coding agents specifically, that resets who gets to build serious agent infrastructure.
Simon Willison shipped condense-json 1.0 to shrink the SQLite logs his LLM tool generates. The unglamorous release is a signal about where the token era's costs actually accumulate: not in the model call, but in everything you keep afterward.
OpenAI's Agents SDK v0.19.0 lets the model write code that calls tools instead of picking them one at a time. The real story isn't the feature. It's what it does to the observability layer other vendors are quietly trying to own.
Black Forest Labs shipped FLUX 3 with a companion video-action robotics model. The story isn't the benchmarks. It's that video generation and video-conditioned control just moved toward commodity, and that changes what your agent can act on.
Muse Spark 1.1 is Meta's first model with a price tag and closed weights. Read past the hypocrisy takes: the fight was never about licensing. It's about who can deploy agents cheapest.
GPT 5.6 ships in three sizes and Codex folds into ChatGPT. Read together, the two moves signal a shift from selling models to selling agent infrastructure, and the distinction matters for anyone running agents daily.
OpenAI killed the model picker to simplify AI, then shipped extra options that confuse people anyway. The lesson for agent operators: the routing layer is the new control surface, and hiding it doesn't make it disappear.
Meta's Muse Spark 1.1 is the first Spark model with an API, and it leads with tool calling and computer use, not benchmarks. The tell isn't the model. It's that Meta wants you building agents that act.
