THINKCAR Launches Tyler AI Diagnostic Agent with Transparent Fault Analysis
THINKCAR has unveiled Tyler, an artificial intelligence diagnostic agent designed to provide transparent, evidence-based vehicle fault analysis on its T394 AI tablet. The system represents a departure from AI diagnostic tools that deliver single, unverified answers by instead presenting ranked possibilities with cited sources.
The Shenzhen-based company announced that Tyler cross-references two independent data layers: OEM repair databases and official recall alerts, safety investigations, complaints, and vehicle safety ratings. When analyzing fault codes, the system generates a prioritized list of probable causes, each accompanied by diagnostic trouble codes and recommended inspection sequences.
Performance and Limitations
According to THINKCAR's internal testing, Tyler achieves fault-diagnosis accuracy exceeding 95 percent and fault-prediction accuracy exceeding 85 percent on covered vehicle models. The company emphasizes that these figures improve as the system's data accumulates, but stresses that Tyler functions as a diagnostic aid rather than a replacement for technician judgment.
The AI agent outputs possible fault causes with verification procedures, requiring technicians to confirm findings against the actual vehicle. For instance, a cylinder-1 misfire code such as P0301 can stem from multiple sources including spark plugs, ignition coils, fuel injectors, or compression issues. Tyler organizes these possibilities and provides verification steps, but leaves final confirmation to the technician.
Technical Foundation
Tyler operates on ThinkMind, THINKCAR's proprietary automotive diagnostic large model built on a Multi-Agent architecture. The knowledge base draws from over 100 million diagnostic data records and cases within a global diagnostic ecosystem spanning 215 countries and regions, serving 2.4 million users based on company data.
The system's design aligns with recent peer-reviewed research in AI-based automotive diagnostics. A 2025 study in MDPI Electronics reported 96.8 percent fault-identification accuracy using explainable-AI techniques for multiple-fault diagnosis in automotive electronic systems. A separate 2026 study in Nature's Scientific Reports documented 99.15 percent accuracy for AI-based fault identification on new-energy-vehicle telemetry data.
Functional Capabilities
On the T394 AI diagnostic tablet, Tyler consolidates ten capabilities including multimodal interaction, multi-intent parsing and task orchestration, one-tap AI diagnosis, deep fault analysis, repair-step guidance with smart linking, dual-screen AI diagnosis, fault prediction, customer management with smart alerts, low-barrier smart diagnosis through symptom interviews and dashboard warning-light recognition, and vehicle value-added services including AI vehicle valuation and maintenance guidance.
The system distinguishes itself through ThinkClaw's multi-intent task orchestration, which allows Tyler to execute complex instructions autonomously. A single spoken command can trigger multiple actions such as adjusting screen brightness, running system scans, and generating reports, moving beyond simple question-and-answer functionality.
Voice wake-up and speech recognition debut in English, with additional languages planned for future releases. Every Tyler output requires technician review before action, maintaining human oversight throughout the diagnostic process.
Market Context
THINKCAR positions Tyler as a solution to a growing problem in AI automotive diagnostics: tools that present confident single answers without supporting evidence or source citations. The company notes that incorrect initial diagnoses lead to unnecessary part replacements, customer returns, and erosion of shop credibility.
The T394 AI tablet with Tyler will be available through authorized dealers. THINKCAR separately markets a more affordable AI diagnostic lineup—the THINKTOOL 689BT PRO and MUCAR 892BT PRO—through its online platform mythinkcar.com.
Founded in 2019, THINKCAR operates with AI patents and a nationally registered automotive AI algorithm. Its product ecosystem encompasses eight categories including diagnostic tools, TPMS, ADAS calibration, EV diagnostics, and remote service platforms.
