World's No.1 Open-Source Agent in Portal Automation—Transforming Supply Chains at 1/14th the Cost

Ventus, in collaboration with UNC, co-developed an AI agent specializing in web automation, achieving top performance in research benchmarks for web navigation and automation among open-source solutions.

In real-world tasks—such as data entry into TMS and other portals, Ventus' agent outperforms significantly. It completes tasks at 1/14th the cost and 1/11th the time compared to frontier models like OpenAI CUA.

Accurate, efficient and secure for the real world

Frontier labs like OpenAI are advancing rapidly in developing larger, more powerful models with reasoning capabilities akin to "super intelligence" (e.g., the IQ of a Math PhD).

However, in real-world scenarios, these models are:

1. overkill, expensive to operate - most tasks don't need super intelligence,
2. require all the data to be sent online,
3. closed-source, making them opaque - requiring users to rely on a black box for their crucial operations.

Addressing these concerns is crucial if we aim to deploy these models for critical tasks within the supply chain effectively.

Ventus is adopting a unique approach by prioritizing models that are both efficient and secure. Our solutions deliver competitive performance compared to closed-source options like OpenAI, while significantly outperforming other open-source alternatives.

We also developed a mode that keeps all sensitive data local instead of sending all of it to the cloud - it almost doubled the accuracy over previous results.

This innovation is particularly impactful for industries such as supply chain management, freight, and logistics where privacy and efficiency are paramount. Smaller models have the advantage of running locally, providing a strong filter for sensitive information and significantly improving data privacy. Moreover, their reduced size and minimal resource requirements make automation more accessible and cost-effective.

January 2026 Update

Since late 2025, two shifts have materially changed day-to-day portal work in logistics: widespread adoption of passkeys (WebAuthn) for carrier and terminal logins, and tighter human-verification controls (e.g., Cloudflare Turnstile/hCaptcha) on appointment, billing, and dispute pages. At the same time, the FMC’s demurrage/detention billing rule (effective 2024) is now showing up in enforcement and contracts, with shippers pushing their providers to retain verifiable evidence for every invoice. These changes favor browser-native automation that is device-aware, policy-compliant, and auditable. Our engagement with the intermodal community through Ventus AI Joins IANA to Accelerate AI-Driven Automation in Intermodal Logistics reflects this reality: automation must “speak the same language” as the portals teams already touch every day.

Updated best practices we now recommend: 1) Treat passkeys as assets with lifecycle policies. Pre-register hardware-backed WebAuthn credentials for each portal in a dedicated, SOC 2–controlled browser profile per customer, store attestation metadata, and set rotation reminders aligned to vendor policies. Maintain an MFA fallback matrix (phone call, SMS, authenticator app) and test it quarterly; our agents can handle both passkeys and voice/OTP flows without user intervention. 2) Standardize a dispute “proof pack.” For demurrage/detention, automatically capture timestamped screenshots and page HTML for gate-in/out, appointment confirmations, line notices, and tariff references; extract invoice-required fields (container, billable party, clock start/stop, free-time basis) into a single export. One drayage customer using this pattern reduced average dispute resolution from 9 days to 48 hours and recovered $42,000 in misbilled charges in a single quarter.

The market continues to be browser-first: in Q4 2025, over 80% of logistics automations we deployed executed entirely in portals or email—not APIs—because many TMS/terminal systems still limit external interfaces for scheduling, bids, and billing evidence. That makes reliability under real-world friction (rolling UI changes, new MFA, seasonal traffic) the key success factor. To hedge change risk, we ship every workflow with: monitored CSS/XPath “canaries,” a human-verification fallback policy, and live notifications to Slack or Teams when a portal’s contract or surcharge field changes. We’re sharing field results and implementation playbooks in Announcing the InTek × Ventus Podcast: Real AI Agents, Real Logistics Results and continuing the modernization dialogue started in Digital Transformation in Logistics and Supply Chain: Insights from Peter Q., CEO of Ventus AI, at AGIC 2024.

Action to take this quarter: audit three high-friction portal flows—drayage appointment scheduling, rail billing disputes, and LTL rate/quote capture—and implement the passkey/MFA policy plus proof-pack standard. With browser-native agents that handle MFA, CAPTCHAs, and even phone calls, most teams are live in under seven days and insulated from the next round of portal changes.

Our approach

Our approach involves a "symbiotic" partnership between two types of AI models:

• Large Language Models (LLMs): Highly sophisticated models capable of understanding complex tasks and generating high-quality execution strategies. • Small, Efficient LLMs: These models learn and distill knowledge from their larger counterparts, becoming more efficient, faster, and cost-effective at performing tasks.

Through a novel framework called AgentSymbiotic, our small LLMs not only replicate the success of large LLMs but also explore and discover innovative solutions. By doing so, they enrich the overall intelligence and performance of our large LLMs in a continuous, iterative improvement cycle.

Key innovations include:

• Iterative Learning: Small LLMs learn from the rich, detailed knowledge provided by large LLMs, refining their abilities over time. • Speculative Data Synthesis: Small LLMs proactively explore new ways to solve tasks, allowing the entire system to discover better, more efficient methods, which continuously enhances the capabilities of both LLM types. • Multi-task Learning: Small LLMs don't just copy the actions; they also learn the reasoning behind these actions, significantly improving their decision-making and problem-solving skills. • Hybrid Privacy Mode: Small LLMs are also deployed locally for tasks involving sensitive information, taking on the responsibility to ensure critical data remains private and secure, filtering out sensitive details autonomously.

The applications in logistics and transportation

Logistics and transportation businesses thrive on efficiency and accuracy. Yet, countless hours are lost every day to routine, repetitive tasks—like data entry, manual tracking updates, or complex scheduling. At Ventus, we've developed cutting-edge AI solutions that autonomously integrate with cloud-based portals—even those without APIs—transforming how logistics operations are conducted.

Stay tuned for our next post, where we'll explore how AI agents that mimic human interactions on web portals can bridge fragmented systems and unlock new levels of efficiency in logistics and beyond.

For more details on our approach, please refer to the project website and the technical writeup.