📼 Introducing VCR
Vercel Container Registry stores your team's container images, so you can use them in Functions or Sandboxes.
https://t.co/qowWocVafU
Claude Opus 5 is now on AI Gateway with support for fast mode.
Improvements versus previous Opus models on agentic coding and quality at lower reasoning effort levels.
𝚖𝚘𝚍𝚎𝚕: '𝚊𝚗𝚝𝚑𝚛𝚘𝚙𝚒𝚌/𝚌𝚕𝚊𝚞𝚍𝚎-𝚘𝚙𝚞𝚜-𝟻'
https://t.co/LEUrGqiWeU
You can now configure where Vercel Workflows keeps each run's state, defaulting to where the run starts.
For agents, state can now live in the same region as the loop writing it.
https://t.co/EE2keUi8n4
You can now operate any running Sandbox without leaving the browser.
Run commands, browse files, take snapshots, straight from the dashboard.
https://t.co/B3Qo0KMuJt
Transcribe live audio on AI Gateway.
Audio streams in, the transcript streams back as you speak. Build voice agents with 𝚜𝚝𝚛𝚎𝚊𝚖𝚃𝚛𝚊𝚗𝚜𝚌𝚛𝚒𝚋𝚎 in the AI SDK.
https://t.co/rgnW0BRRAF
Free through 8/3: Ling 3.0 Flash from @AntLingAGI is on AI Gateway.
A MOE model for agentic workflows, with support for thinking and non-thinking modes.
𝚖𝚘𝚍𝚎𝚕: '𝚒𝚗𝚌𝚕𝚞𝚜𝚒𝚘𝚗𝚊𝚒/𝚕𝚒𝚗𝚐-𝟹.𝟶-𝚏𝚕𝚊𝚜𝚑-𝚏𝚛𝚎𝚎'
https://t.co/Yx10OuvU2T
extensions package capabilities that any eve agent can import and use:
▪️ A browser-use extension can include tools for navigating a site
▪️ A memory extension can use hooks to capture context and tools to recall it
▪️ A self-improvement extension can pair hooks with dynamic instructions
Here is how it works 👇
1. Build and publish an extension:
𝚗𝚙𝚡 𝚎𝚟𝚎@𝚕𝚊𝚝𝚎𝚜𝚝 𝚎𝚡𝚝𝚎𝚗𝚜𝚒𝚘𝚗 𝚒𝚗𝚒𝚝 𝚖𝚢-𝚌𝚛𝚖
Define capabilities mirroring the agent directory:
@𝚊𝚌𝚖𝚎/𝚌𝚛𝚖
├── 𝚙𝚊𝚌𝚔𝚊𝚐𝚎.𝚓𝚜𝚘𝚗
└── 𝚎𝚡𝚝𝚎𝚗𝚜𝚒𝚘𝚗/
⠀⠀⠀├── 𝚎𝚡𝚝𝚎𝚗𝚜𝚒𝚘𝚗.𝚝𝚜
⠀⠀⠀├── 𝚝𝚘𝚘𝚕𝚜/𝚜𝚎𝚊𝚛𝚌𝚑.𝚝𝚜
⠀⠀⠀├── 𝚌𝚘𝚗𝚗𝚎𝚌𝚝𝚒𝚘𝚗𝚜/𝚊𝚙𝚒.𝚝𝚜
⠀⠀⠀├── 𝚜𝚔𝚒𝚕𝚕𝚜/𝚝𝚛𝚒𝚊𝚐𝚎/𝚂𝙺𝙸𝙻𝙻.𝚖𝚍
⠀⠀⠀└── 𝚒𝚗𝚜𝚝𝚛𝚞𝚌𝚝𝚒𝚘𝚗𝚜.𝚖𝚍
Run 𝚎𝚟𝚎 𝚎𝚡𝚝𝚎𝚗𝚜𝚒𝚘𝚗 𝚋𝚞𝚒𝚕𝚍 and publish it to npm.
2. Install and mount the extension on any eve agent
𝚗𝚙𝚖 𝚒𝚗𝚜𝚝𝚊𝚕𝚕 @𝚊𝚌𝚖𝚎/𝚌𝚛𝚖
Mount it in agent/extensions:
// 𝚊𝚐𝚎𝚗𝚝/𝚎𝚡𝚝𝚎𝚗𝚜𝚒𝚘𝚗𝚜/𝚌𝚛𝚖.𝚝𝚜
𝚒𝚖𝚙𝚘𝚛𝚝 𝚌𝚛𝚖 𝚏𝚛𝚘𝚖 "@𝚊𝚌𝚖𝚎/𝚌𝚛𝚖";
𝚎𝚡𝚙𝚘𝚛𝚝 𝚍𝚎𝚏𝚊𝚞𝚕𝚝 𝚌𝚛𝚖();
Immutable static assets can now be cached across deployments on the Vercel CDN.
For frequently deployed projects we see:
▪︎ 60%+ global TTFB reduction
▪︎ 30% faster deployments
▪︎ Less usage and improved performance
https://t.co/I5twAhxe7F
Service tiers are now available on AI Gateway.
Use 𝚙𝚛𝚒𝚘𝚛𝚒𝚝𝚢 for the fastest token throughput or 𝚏𝚕𝚎𝚡 for cost-sensitive, latency-tolerant requests. https://t.co/fatZkTh1a2
Laguna S 2.1 from @poolsideai is on AI Gateway and free for a limited time.
Open-weight model that rivals larger models on agentic coding and long runs.
• 𝚙𝚘𝚘𝚕𝚜𝚒𝚍𝚎/𝚕𝚊𝚐𝚞𝚗𝚊-𝚜-𝟸.𝟷-𝚏𝚛𝚎𝚎
• 𝚙𝚘𝚘𝚕𝚜𝚒𝚍𝚎/𝚕𝚊𝚐𝚞𝚗𝚊-𝚜-𝟸.𝟷
https://t.co/F1X8znZDWE
Gemini 3.6 Flash and Gemini 3.5 Flash-Lite are now on AI Gateway.
• 𝚖𝚘𝚍𝚎𝚕: '𝚐𝚘𝚘𝚐𝚕𝚎/𝚐𝚎𝚖𝚒𝚗𝚒-𝟹.𝟼-𝚏𝚕𝚊𝚜𝚑'
• 𝚖𝚘𝚍𝚎𝚕: '𝚐𝚘𝚘𝚐𝚕𝚎/𝚐𝚎𝚖𝚒𝚗𝚒-𝟹.𝟻-𝚏𝚕𝚊𝚜𝚑-𝚕𝚒𝚝𝚎'
https://t.co/m4cgMDZ4Yt
Data transfer into Vercel Sandbox is now free.
▪︎ Clone Git repos and install packages without data charges
▪︎ Pull artifacts or datasets from external sources at no cost
https://t.co/naHqPMSTHu
The Vercel Plugin is now available in the Kimi Code CLI.
Kimi Code can now draw on Vercel platform knowledge on demand, with skills for Next.js, AI SDK, Vercel Functions, and more.
https://t.co/zGj7vUauEg
GLM 5.2 via Novita is 35% off on AI Gateway through July 24.
𝚜𝚝𝚛𝚎𝚊𝚖𝚃𝚎𝚡𝚝({
𝚖𝚘𝚍𝚎𝚕: '𝚣𝚊𝚒/𝚐𝚕𝚖-𝟻.𝟸',
𝚙𝚛𝚘𝚟𝚒𝚍𝚎𝚛𝙾𝚙𝚝𝚒𝚘𝚗𝚜: {
𝚐𝚊𝚝𝚎𝚠𝚊𝚢: { 𝚘𝚛𝚍𝚎𝚛: ['𝚗𝚘𝚟𝚒𝚝𝚊'] }
}
})
https://t.co/swwtZa713D
GitHub Tools now mints short-lived, scoped tokens via Vercel Connect, so your agent doesn't have to juggle API keys.
Presets like code-review map to Connect scopes.
https://t.co/x54Gct7hJb
You can now use Vercel Plugin with @GitHub Copilot CLI and VS @Code:
• 30 platform skills
• 3 specialized agents
• 4 built-in commands
• Real-time code validation
Get started ↓
https://t.co/uUsPOiuv3Z
Kimi K3 is now on Vercel AI Gateway.
Moonshot's newest model, now with 1M context and native vision.
𝚖𝚘𝚍𝚎𝚕: '𝚖𝚘𝚘𝚗𝚜𝚑𝚘𝚝𝚊𝚒/𝚔𝚒𝚖𝚒-𝚔𝟹'
https://t.co/An3cMkmOTN