Menlo Ventures：2026 年企业 AI 市场报告
原题Menlo Ventures: 2026 Enterprise AI Market Report
推荐理由产品与投资人可用的数字集：企业 AI 部署支出一年涨 5.4 倍至 1770 亿美元，OpenRouter 智能体流量已占 token 总量八成。
一句话
Menlo Ventures 的 2026 年《企业 AI 状况报告》揭示了 AI 部署支出的爆发式增长、智能体驱动的 token 消耗，以及企业 AI 采纳格局的重大转变。
完整摘要
根据 Menlo Ventures 的《2026: The State of Enterprise AI》报告，企业 AI 部署支出从 2025 年的 330 亿美元激增至 2026 年 9 月的 1770 亿美元，增长 5.4 倍，超过去年增长 3.9 倍的速度，已占公共云收入的 30% 以上。智能体（Agents）是推动增长的主要力量：在 OpenRouter 上，2026 年 2 月智能体流量超过人类流量，随后相较于人类增长 37 倍，最终占 token 总量的 80%，或每周 131 万亿。企业预算计划增至 2025 年水平的 1.9 倍，实际平均增至 4.7 倍，因平行智能体能够随工作量增长而扩张支出，而非随员工人数；中位数仅达 2.2 倍，说明少数重度用户推动了整体消耗的大部分增长。Anthropic、OpenAI 和 Google 合计占 deployment dollars 的 69%，即 1230 亿美元，较去年的 50% 显著提升；Anthropic 在企业 API 支出中份额从 40% 上升至 55%，OpenAI 从 27% 上升至 31%，Google 从 21% 下降至 9%。买家持续为能力付费：前沿模型支出份额从 35% 上升至 63%，最流行的模型从 Claude Sonnet 4.5 和 GPT-5 转向更昂贵的 Claude Opus 5.5 和 GPT-6 Astra。企业每为构建模型支出 1.7 美元，就花 1 美元用于 AI，总计 1770 亿美元 versus 1025 亿美元；仅 Anthropic 和 OpenAI 就占据 AWS、Azure 和 Google Cloud 收入的 25% 以上。编码和聊天机器人构成 773 亿美元应用市场的 76%，因编译器和测试工具使智能体能够验证结果；编码领域突破 340 亿美元，Claude Code 和 Codex 合计 240 亿美元，占比 70%。开源模型处理 token 比例从 11% 上升至 19%，但支出份额仅 4%，因其平均价格下降 3.3 倍至每百万美元 0.13 美元；中国模型在开源支出中占据 89%，以 Kimi 为首，份额高达 61%。AI 正在扩大软件预算：新增 AI 支出的 55% 来自新建预算，仅 9% 或 126 亿美元从软件转向 AI-native 供应商，使应用 SaaS 增长率从 13% 降至 11%。AI-native 供应商在应用支出中占比提升至 77%（原 63%），编码领域为 96%；而传统厂商在 IT 领域仍 hold 有 91%，在 HR 领域为 68%，AI 必须融入现有记录系统中。
According to Reuters, OpenAI stored most of its code in one monorepo, so anyone with code access could read details of many projects, including sensitive ones unrelated to their own work. OpenAI staff warned leaders for years that anyone with code access could see details of many projects. Those complaints reached executives and the board, yet some staff concluded that keeping up with Google and Anthropic came first. Although, OpenAI says model weights and much of its most sensitive work sit outside the monorepo. > According to @MenloVentures 「2026: The State of Enterprise AI」 > Enterprise AI deployment spend jumped from $33B in 2025 to $177B by Sept-2026, a 5.4x rise that beat the prior year's 3.9x growth and now equals over 30% of public cloud revenue. > Agents drove the jump: on OpenRouter, agent traffic passed human traffic on Feb-2026, then grew 37x versus 5x for humans, reaching 80% of tokens, or 131T per week. > Companies budgeted 1.9x their 2025 AI spend but averaged 4.7x, since parallel agents scale spending with work rather than headcount; the median reached only 2.2x, so a few power users drive a disproportionate share. > Anthropic, OpenAI and Google took 69% of deployment dollars, $123B, up from 50%; Anthropic's share of enterprise API spend rose from 40% to 55%, OpenAI's from 27% to 31%, Google's fell from 21% to 9%. > Buyers keep paying for capability: frontier models' spend share rose from 35% to 63%, and the most popular models shifted from Claude Sonnet 4.5 and GPT-5 to pricier Claude Opus 5.5 and GPT-6 Astra. > Enterprises spend $1.70 using AI for every $1 spent building models, $177B versus $102.5B, and Anthropic and OpenAI alone account for an estimated 25%+ of AWS, Azure and Google Cloud revenue. > Coding and chatbots make up 76% of the $77.3B app market, since compilers and tests let agents verify results; coding hit $34.0B, with Claude Code and Codex taking $24.0B, or 70%. > Open-source handles 19% of tokens, up from 11%, but only 4% of spend, as its average price fell 3.3x to $0.13 per million; Chinese models take 89% of open-source spend, led by Kimi at 61%. > AI is expanding software budgets: 55% of incremental AI spend came from new budgets, and only 9%, or $12.6B, shifted from software to AI-native vendors, trimming application SaaS growth from 13% to 11%. > AI-native providers took 77% of application spend, up from 63%, and 96% of coding, but incumbents hold IT at 91% and HR at 68%, where AI must fit existing systems of record.
A solid study from @MenloVentures
"2026: The State of Enterprise AI"
> Enterprise AI deployment spend jumped from $33B in 2025 to $177B by Sept-2026, a 5.4x rise that beat the prior year’s 3.9x growth and now equals over 30% of public cloud revenue.
> Agents drove the jump: on OpenRouter, agent traffic passed human traffic on Feb-2026, then grew 37x versus 5x for humans, reaching 80% of tokens, or 131T per week.
> Companies budgeted 1.9x their 2025 AI spend but averaged 4.7x, since parallel agents scale spending with work rather than headcount; the median reached only 2.2x, so a few power users drive a disproportionate share.
> Anthropic, OpenAI and Google took 69% of deployment dollars, $123B, up from 50%; Anthropic’s share of enterprise API spend rose from 40% to 55%, OpenAI’s from 27% to 31%, Google’s fell from 21% to 9%.
> Buyers keep paying for capability: frontier models’ spend share rose from 35% to 63%, and the most popular models shifted from Claude Sonnet 4.5 and GPT-5 to pricier Claude Opus 5.5 and GPT-6 Astra.
> Enterprises spend $1.70 using AI for every $1 spent building models, $177B versus $102.5B, and Anthropic and OpenAI alone account for an estimated 25%+ of AWS, Azure and Google Cloud revenue.
> Coding and chatbots make up 76% of the $77.3B app market, since compilers and tests let agents verify results; coding hit $34.0B, with Claude Code and Codex taking $24.0B, or 70%.
> Open-source handles 19% of tokens, up from 11%, but only 4% of spend, as its average price fell 3.3x to $0.13 per million; Chinese models take 89% of open-source spend, led by Kimi at 61%.
> AI is expanding software budgets: 55% of incremental AI spend came from new budgets, and only 9%, or $12.6B, shifted from software to AI-native vendors, trimming application SaaS growth from 13% to 11%.
> AI-native providers took 77% of application spend, up from 63%, and 96% of coding, but incumbents hold IT at 91% and HR at 68%, where AI must fit existing systems of record.
- 回复
- 46
- 转发
- 10
- 点赞
- 54
- 浏览
- 6082