PPIO Launches AgenticCloud
智东西 · wechat · 2026-07-19
At WAIC 2026, PPIO launched AgenticCloud and an intelligent model gateway designed for the Agent era, branded as an "intelligent Token factory." The company notes that to handle long-chain reasoning, multi-step logic, and high-frequency tool calls from Agents, cloud services must shift from being "human-centric" to "Agent-centric," prioritizing low latency, high throughput, cost-efficiency, and stability.
Their solution includes using MoM (Mixture of Models) for multi-model fusion and smart routing. Inference costs are reduced through context compression, compute reuse, budget guardrails, and fallback mechanisms. The inference engine and PromptCache are optimized to boost efficiency for high-frequency tool calls and long contexts. Furthermore, the Agent sandbox is built on microVMs, achieving a 200ms cold start, system-level isolation, and auto-pause/resume. It is compatible with E2B interfaces and has seen its business scale grow over 123x in less than a year.
The Harness layer integrates BrowserUse, ComputerUse, CodeInterpreter, and MCPServer, covering browser automation, desktop control, code execution, and external service integration. It is also compatible with frameworks like LangChain, CrewAI, and AutoGen. PPIO concluded with its growth metrics: as of June 2026, daily Token calls exceeded 1.2 trillion, AI cloud revenue grew more than 10x year-over-year, and average GPU utilization remained above 75%.
More from coding & agent
- Warp's six non-engineering teams all run on Linear and Claude Code — mon__lim · 2026-09-11
- Is inference latency becoming the biggest bottleneck for production AI agents? — Euphoric_Sea632 · 2026-09-11
- Anthropic researcher: 99% of engineers now run swarms of 300+ self-improving agents — AlishaOutridge · 2026-09-11
- Gergely Orosz: Shipping 10x PRs With AI Agents, Sites Fill With Small Regressions — ducha_aiki · 2026-09-11
- Same Echo Maze prompt, three frontier models: all passed visually but shipped the same hidden bug — eyishazyer · 2026-09-11
- Astra storyboards plus Minimax H3 per-shot generation boost video success rates — Hailuo_AI · 2026-09-11