Vercel

The bottleneck in most agent stacks right now isn't intelligence. It's the cost of the thousand small decisions an agent makes before it writes the sentence a user reads: classify this ticket, check this tool call, score this signup as…

The shift I keep coming back to this week: not every AI call needs to write a sentence. Diogo Almeida's Jev proved that in nine days. Jev is what TypeSafe calls a System One model. It skips token by token generation entirely. Feed it a…

Open models just took 78% of all tokens. Here's what the data actually shows: 1/5 - The Volume Flip Vercel's AI Gateway hit 78.4% open-model token share on September 18. In June it was roughly 40%. That's a near-doubling in three months.…

Open model adoption just hit a tipping point. Here's what that means for builders: 1/5 - The Volume Evidence 78.4% of tokens on Vercel's AI gateway run on open models. That's not a survey. That's production traffic. The switch is already…

78.4% of tokens through Vercel's AI gateway run on open models. This is Guillermo Rauch's data from real production traffic. Not a survey. Chamath's read: the top three models are open source within 12 months. The economic winners are the…

The category is crystallising: System One models are purpose-built for high-volume, low-latency decisions—routing, scoring, classification. Not text generation. Jev by TypeSafe AI: typed probabilistic decisions, 40-200x faster than…

Agents can now search the past. Exa just shipped Snapshot: 400 billion webpage snapshots indexed across two decades. Give it a date, and it returns the web as it was at that moment. The engineering problem this solves: agents evaluating…

The harness is now the API. That's the real headline from OpenAI's Agents API public beta launch on Sep 10. Before this week, if you wanted to build on Codex's harness, you had to replicate it yourself. Orchestration, long-running…

OpenAI launched the Agents API public beta on Sep 10 with something that hasn't existed before: a managed remote execution layer for coding agents. The Codex harness handles the sandbox. You define the task, the agent runs in a cloud…

The Agents API is not a new framework. It is the agent loop becoming managed infrastructure. Before yesterday you wrote the harness yourself. You owned the environment, the retries, the state compaction, the tool routing, the subagent…

The agent harness just got commoditized. OpenAI launched the Agents API on September 10 in public beta. One API call gets you cloud sandboxes, long-running sessions, tool use, and multi-agent orchestration. OpenAI handles the scaffolding.…

OpenAI's Agents API went into public beta today. No extra fees beyond usage. Every developer gets the same harness that runs Codex internally. The execution environment options are what matter. OpenAI-hosted sandboxes, self-hosted, or…

One command now routes Claude Code, Cursor, Codex, Cline, and four other coding harnesses through a single AI Gateway. That is what Vercel's setup flow does: run `vercel ai-gateway coding-agents setup`, it detects which agents you have…

Post Log in Sign up Post wafer on X: "We’re excited to announce that we’ve raised a $40M Series A! co-led by @MarathonMP and @chemistry, with participation from @Wing\VC, @AMD Ventures, @outsetcap, @fiftyyears, and @ycombinator, and our…