Hey Lobsters, welcome to this new edition of ShanghAI.

Every Enterprise CTO who looked at Chinese AI models last year hit the exact same brick wall: procurement approvals, new vendor security reviews, legal sign-offs, and separate billing accounts.

That wall just collapsed.

OpenAI Enterprise and Codex users can now run Moonshot AI's Kimi K3 directly through their existing OpenAI spending commitments, served via US inference partner Baseten.

No new vendor approval. No new master service agreement. No separate invoice.

Kimi K3 isn't a lightweight open model, it packs 2.8 trillion parameters with a 1-million-token context window. While it sits just under top-tier frontier models like GPT-5 on extreme reasoning edge cases, it routinely matches or beats previous-gen models across long-context coding, repository analysis, and document processing.

At $3/M input and $15/M output tokens (with cached inputs at just $0.30/M), it brings high-end capacity at a fraction of standard enterprise tier rates.

The math for business leaders is simple: you aren't choosing between "Enterprise Compliance" or "Cost Arbitrage." You are routing heavy workloads through your existing, pre-approved OpenAI commitment while American infrastructure handles the execution.

This year AI shook up creative strategy. Most teams are still working like it didn't.

Ari Murray, Chief Digital Officer at Salt and Stone sits down with Bryan Harvey, Head of Brand at Hightouch, for an honest conversation about what's breaking on creative teams right now, how the roles are changing, and what a day looks like when creatives spend it on ideas, concepts, and craft while agents handle the manual execution.

Leave with new insights from 3 experts in the field and 5 prompts to instantly improve your creative output.

Reserve your seat

This arrangement marks the first time a Chinese open model sits directly inside OpenAI's enterprise billing environment.

The enterprise market is moving toward single-bill multi-model routing.

Here is how the setup operates for business teams:

1. Unified Corporate Billing

Your company does not sign a new contract with Moonshot AI. Usage draws down from the pre-approved annual OpenAI Enterprise spend your CFO already signed off on.

2. Managed US Infrastructure & Compliance

The model is served on US-based server clusters operated by Baseten. Prompts operate under strict Zero Data Retention (ZDR) rules, meaning your internal corporate code, customer inboxes, or legal documents are never stored or used for retraining.

3. Plug-and-Play Developer Workflows

Because it connects through OpenAI-compatible interfaces (like Codex and the Responses API), engineering teams don't need to rebuild internal apps or retrain employees. Swapping models is as simple as selecting a dropdown or changing a single setting lin

Every agent handling refunds, account changes, or tickets needs three answers: what it can pick up, what it's allowed to do, and when it hands back. Running Agents in Customer Work is four conversations on agentic AI in customer ops. Register now, four Tuesdays, 10 a.m. PT.

Register now

This non-technical playbook shows how business and engineering leads can optimize their OpenAI Enterprise commitment by routing routine, heavy tasks to Kimi K3.

Sit down with your engineering or operations managers and review your highest-volume AI tasks. Look for:

Have your admin or tech lead navigate to your existing OpenAI Enterprise / Codex admin panel. Select Kimi K3 from the available model options provided via the Baseten integration.

No procurement ticket, credit card entry, or legal review is required.

Confirm with your IT admin that Zero Data Retention (ZDR) is active on the deployment. This ensures all data processed through the model stays strictly within US-managed infrastructure and is discarded immediately after generating the answer.

Track your OpenAI commitment consumption over 30 days. Because routine long-context processing consumes significantly fewer commitment credits on Kimi K3, your overall enterprise allocation will stretch substantially further across departments.

You don't need a new software budget or a complex IT migration. You simply start spending your existing OpenAI commitment more efficiently.

If your leadership team wants to audit your enterprise AI spending, review multi-model routing options, or stretch your current cloud budget:

Book an Architecture & Cost-Audit Session with LeanAI