Kimi for Financial Services: Moonshot Sells Vertical Research at $7–$104/Month, Wired via MCP to S&P Global and Wind

Kimi for Financial Services: Moonshot Sells Vertical Research at $7–$104/Month, Wired via MCP to S&P Global and Wind

On September 17, 2026, Beijing-based Moonshot AI launched Kimi for Financial Services — a full vertical research product from a frontier-model lab, with institutional data partnerships, packaged workflow "Skills," and compliance controls, priced at consumer-band rates. It is the most aggressive data point yet on vertical research pricing, and it extends the financial-research vertical thread (Sacra: Premium Private Market Research as Agent-Ready Infrastructure, Institutional MCP Financial Data Servers: FactSet, S&P Global, and PitchBook Unbundle) into direct price competition.

What was announced. CNBC: "Kimi users can directly access information commonly used for analysis and reports, thanks to a number of industry data partners such as S&P Global Market Intelligence, Crunchbase, Wind, local financial news leaders and business database Tianyancha, according to Moonshot." Moonshot also says Kimi "can also directly access the U.S. Securities and Exchange Commission's EDGAR system for public companies' financial filings, the IMF, World Bank and the U.S. Federal Reserve Economic Data site (FRED)" (CNBC, Sep 17 2026). Named users include investment bank CICC and Sequoia China (now Hong Shan).

The price point is the story:

"Subscriptions to Kimi start at 49 yuan ($7.31) a month, and can go up to 699 yuan ($104.23)."¹ — CNBC

That top tier ($104/mo) undercuts Western enterprise research seats by an order of magnitude — compare Elicit's $169 Scale tier in Elicit's Two-Front Expansion: Multiplayer Research Workspace + Evidence API/MCP Backend or institutional data workflows. The product rides Kimi K3, Moonshot's July 2026 open-weight model that it claims outperforms all rivals except Claude Fable 5 and GPT-5.6 overall; the company has reportedly filed confidentially for a Hong Kong IPO.

The architecture, from Moonshot's own announcement (kimi.com, Sep 17 2026): the solution bundles "10+ 权威数据源、9 项金融技能（Skill）、5 项合规安全措施，以及机构级的数据建模和报告交付能力" (10+ authoritative data sources, 9 financial Skills, 5 compliance/security measures, plus institutional-grade data modeling and report delivery²). The data plumbing is MCP-native:

"这些数据通过 MCP 直接接入研究任务，关键数字可以回到原始出处核对，来源、口径清晰可查。" — "This data connects directly into research tasks via MCP; key figures can be checked against the original source, with provenance and methodology clearly traceable." — kimi.com announcement

The 9 Skills are named deliverable workflows — 机构财务建模 (institutional financial modeling), 研究报告 (research reports), PPT, 金融动态图表 (dynamic financial charts), 业绩点评 (earnings commentary), 一致预期地图 (consensus maps), 组合复盘 (portfolio review), 持仓早报 (holdings morning report), HK IPO 透镜 (HK IPO lens). Products span kimi.com, the Kimi Work desktop app, and Kimi Code, with a browser extension (formerly Kimi WebBridge) and Computer Use so "systems without APIs can still join the workflow," plus Cloudflare developer tooling exposing 2,500+ official API endpoints. For compliance, Moonshot is co-building a "风险评估网关" (risk-assessment gateway) with CITIC Construction Investment Securities (中信建投证券), implementing five measures: data classification tiers, PII minimization/filtering, per-call data-source and tool authorization, output provenance with mandatory human review ("初步材料经业务人员复核，不直接形成立项结论" — preliminary materials are reviewed by business staff and do not directly form conclusions), and full audit traceability.

Deutsche Bank's Beijing branch manager Samuel Fischer appears in Moonshot's promo: "The real inflection point really is the combination of stronger AI capabilities with professional expertise" (CNBC; the bank declined to comment on whether it is a client).

What this means for the Hey, Lefty map

• A frontier-model lab has executed the full vertical playbook — proprietary data partnerships (via MCP), packaged Skills, compliance gateway, deliverable outputs, named institutional clients — at $7–$104/month. This is the same playbook Sacra and the institutional MCP vendors run, at a fraction of the price, from an open-weight-model cost base.
• The two primitives Kimi uses are the two primitives Hey, Lefty's layer uses: MCP for data ingestion, Skills for packaged workflows. The market's building blocks are standardizing fast; differentiation shifts to evidence standards, audit, and orchestration quality — not access.
• Watch the geography. S&P Global Market Intelligence and Crunchbase are Western data partners already inside a Chinese lab's product. If this goes international (English product, more Western clients), it becomes direct price competition in every financial-research segment.