UBTECH Thinker

Thinker 1.0 · Family: Thinker

UBTECH foundation model for embodied intelligence (Thinker 1.0, H1 2026); per the company small parameters, high performance, ranked first on nine embodied-AI leaderboards.

✓ Active✓ Public access⚖ Open sourceRobotics foundation model📁 Thinker

🏢UBTECH RoboticsProducer

Deployment:📱 On-device☁ Cloud

Overview

UBTECH Thinker is the base foundation model for embodied intelligence developed by UBTECH Robotics. It is the first tier of the three-layer Thinker stack (foundation → world model → action model) on which the company builds the cognitive capabilities of its humanoids.

Characteristics

The first generation (Thinker 1.0) was released in the first half of 2026. According to UBTECH the model is characterized by a small parameter count with high performance and is fully open source (company claim). It ranked first on nine embodied-intelligence leaderboards. The exact architecture, parameter count and license have not been publicly disclosed.

Classification

Robotics foundation model

Family: Thinker

Applications

Robotic manipulation Robot policy training

Access & deployment

On-deviceCloud

Weights: Open source

Key parameters

📥 Input: text, image, robot sensors

Robotics

Embodied task planningScene understandingSpatial reasoningRobot control

Technical specification

License

Open source (deklaracja UBTECH; brak potwierdzonego publicznego repozytorium)

Hardware requirements

Not publicly disclosed. The model targets the cognitive layer of UBTECH humanoids (on-device and cloud inference).

Modalities

⬇ Input

textimagerobot_sensors

⬆ Output

robot_actionstext

Capabilities and applications

Native model capabilities

Advanced reasoning

The ability to perform multi-step, structured reasoning: analysing problems, planning steps, and drawing conclusions from hypotheses. Reasoning-first models (e.g. GPT-5.1 Thinking) dedicate a portion of inference to chains of thought before responding.

Category: reasoning

Planning

Forming and executing action plans for complex tasks.

Category: planning

Multimodal understanding

Category: multimodal

Multi-step reasoning

Carrying out multi-step chains of reasoning across long, complex tasks.

Category: reasoning

Robotics

Embodied task planningScene understandingSpatial reasoningRobot control

Application domains