Shizhuan AI 实在智能 (实在Agent)
Turns one sentence into a running RPA workflow, no API required
Overview
Shizhuan AI pairs its TARS language model with ISSUT screen-reading tech so its Agent can operate desktop software by clicking, typing and reading, without needing an API. Used by 5,000+ Chinese enterprises for finance, e-commerce and back-office automation.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 12 Jul 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- Single-workflow builder
- No coding required
- Cloud-hosted agents
- Per-seat or concurrent licensing
- One-time license + implementation fee
- Annual maintenance
- Data stays on-premises
What you can produce with Shizhuan AI 实在智能 (实在Agent)
- Generates RPA workflows from a single natural-language instruction
- ISSUT screen-semantic understanding drives desktop software without APIs
- Self-developed TARS model for task planning and reasoning
- Free community edition for individual no-code automations
- Private/on-premises deployment for regulated industries
- 5,000+ enterprise customers across e-commerce, finance, telecom, government
- Listed in CAICT's 2025 national digital-transformation product catalogue (Mar 2026)
ASEAN Perspective
Shizhuan AI 实在智能 (实在Agent) in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
Shizhuan's pitch is unusually literal: describe a task in one sentence, and the agent reads the screen the way a person does, then executes across whatever software is open, no API integration required. That combination of a purpose-built TARS model with its ISSUT screen-understanding layer is genuinely differentiated in a market full of prompt-only agent wrappers. Case studies back it up with real numbers — Zhouheiya cut monthly reconciliation work from 600 hours to an automated daily run, Jiangzhong Pharmaceutical cut data-retrieval time sixfold. The catch: this is a China-market product through and through, sold in RMB with Chinese-language documentation and support, and enterprise pricing is opaque outside sales conversations. A free community edition exists for simple flows. Worth a look for China-based back-office teams; ASEAN buyers should expect friction on language and support before deployment.
What people say
Shizhuan Intelligent Technology (实在智能), headquartered in Anji, Zhejiang after a 2024 rename from its original Hangzhou entity, builds its Agent product around three self-developed pieces: the TARS large language model for task planning, ISSUT screen-semantic understanding for reading software UIs the way a person would, and its own RPA execution layer. The sales pitch — "you speak, the PC does" — is backed by a March 2026 certification from China's Academy of Information and Communications Technology, which listed Shizhuan Agent in its national digital-transformation product catalogue.
Customer numbers are large for the category: over 5,000 enterprise accounts as of 2025, spanning e-commerce, telecom, energy, finance and government, with Shenzhen Airlines among the named clients. Case studies published by the company carry specific figures rather than vague percentages. Jiangzhong Pharmaceutical's data-retrieval workflows sped up sixfold across more than 50 recurring tasks. A cross-border apparel seller cut multi-platform product-listing time from two hours to five minutes per SKU. Duck-snack retailer Zhouheiya used the Agent to reconcile roughly 90 systems across 50 sales platforms, replacing 600 hours of monthly manual finance work with an automated run of about 14 hours a day — a claimed annual saving near ¥1.88 million and eight finance headcount.
Third-party coverage is thinner than the company's own case studies but not absent. Research house Zhiqi Think Tank's 2025 Q3 AI100 ranking credited the product with a 91% task-completion rate in complex scenarios and framed it as moving "from process execution to goal-driven" automation — a step up from scripted RPA. Coverage in Sohu and elsewhere attributes user trust to the RPA lineage: because the agent operates through the same GUI permissions a human employee would have, it doesn't need brittle API integrations that break when vendor software updates.
Pricing follows a familiar enterprise-software split: a permanently free community edition for individual, no-code automations; a SaaS subscription billed per seat or concurrent license for mid-size teams; and a private-deployment option (license plus implementation plus annual maintenance) aimed at regulated industries like government, finance and pharma that won't put data in the cloud. None of the three has a public price list — quotes come through sales.
Summary of public user & expert reviews, compiled by RECATOOLS.
About this listing
This entry was compiled from publicly available data including Shizhuan AI 实在智能 (实在Agent)'s official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Shizhuan AI 实在智能 (实在Agent) unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Shizhuan AI 实在智能 (实在Agent) directly →
Spotted something out of date? Suggest an update →
Shizhuan AI 实在智能 (实在Agent) in the news
Alternatives to Shizhuan AI 实在智能 (实在Agent)
More in Agents & Automation
