Skip to main content

Turing Begins Joint Research with Automotive Industry Partners on an In-Vehicle E2E Autonomous Driving System and a Physical Foundation Model – Conducted as part of GENIAC, a national project led by METI and NEDO

Turing Inc. (Headquarters: Heiwajima, Ota-ku, Tokyo; CEO: Issei Yamamoto; hereinafter “Turing”), together with Subaru Corporation (Headquarters: Shibuya-ku, Tokyo; Representative Director, President and CEO: Atsushi Osaki; hereinafter “Subaru”) and DENSO CORPORATION (Headquarters: Kariya, Aichi; Representative Member of the Board, President & CEO: Shinnosuke Hayashi; hereinafter “DENSO”), announces that it will begin joint research toward the social implementation of fully autonomous driving—namely, research toward the implementation of an in-vehicle End-to-End (hereinafter “E2E”) autonomous driving system and a physical foundation model.

In this joint research, (1) Turing and Subaru will work on research toward the implementation of an in-vehicle E2E autonomous driving system, and (2) Turing and DENSO will work on the development of a physical foundation model. By bringing together the expertise of each company, Turing will build an integrated framework spanning the entire process from AI model research to implementation in actual vehicles.

Background

The social implementation of fully autonomous driving requires a development framework that cyclically advances both the development of a physical foundation model—which performs advanced recognition and judgment based on large-scale driving data—and the work of installing and validating that model in actual vehicles. To date, Turing has advanced the in-house development of E2E autonomous driving and physical foundation models. Through this joint research, in collaboration with two companies that have long supported the automotive industry, Turing aims to build an integrated framework that spans from research to in-vehicle implementation and public-road demonstration.

Initiatives

This joint research advances the following two initiatives in parallel.

1.In-Vehicle E2E Autonomous Driving System

Together with Subaru, Turing will work on integrating a lightweight E2E autonomous driving model with vehicle control systems, as well as on research with a view to future public-road demonstration. The companies will advance a multi-camera lightweight E2E model and its in-vehicle adaptation, building and validating an autonomous driving system that extends through to vehicle control.

2.Physical Foundation Model

Together with DENSO, Turing will work on the development of a physical foundation model adapted to autonomous driving, as well as its adaptation to actual vehicles. Based on vision-language models and vision state-space models, the companies will advance the technology development needed to acquire linguistic and spatio-temporal understanding from camera sensors alone.

Overview Diagram of Initiative

Future Outlook

Through this joint research, Turing will accelerate the research of E2E autonomous driving systems and physical foundation models, aiming to build a development framework that cyclically connects deployment in actual vehicles, validation in driving environments, and model improvement based on the driving data obtained. In doing so, Turing will establish a foundation for the social implementation of fully autonomous driving, while extending this work beyond autonomous driving toward the establishment of physical AI technologies applicable to a wide range of domains—including multimodal foundation models with an eye toward AI robots.

About This Initiative

This initiative is conducted as part of the Generative AI Accelerator Challenge (GENIAC), a national project led by Japan’s Ministry of Economy, Trade and Industry (METI) and the New Energy and Industrial Technology Development Organization (NEDO), under the “Development of Competitive Generative AI Foundation Models” theme of the Post-5G Information and Communication Systems Infrastructure Enhancement R&D Project.
Selection results: https://www.meti.go.jp/english/press/2026/0514_001.html

About GENIAC

GENIAC (Generative AI Accelerator Challenge) is a national project led by METI and NEDO to strengthen Japan’s capability to develop generative AI. The program primarily provides high-performance computational resources to support the development of foundation models—the core technology behind generative AI—and supports proof-of-concept investigations into the practical application of data and AI.
URL: https://www.meti.go.jp/english/policy/mono_info_service/geniac/index.html

About Turing

Turing is a startup aiming to realize fully autonomous driving. The company develops an End-to-End (E2E) autonomous driving system that handles everything from perception, judgment, and vehicle control based on information from cameras. Turing is also developing a physical foundation model that linguistically understands pedestrians, signs, signals, road conditions, and more—responding flexibly to complex driving scenes. From model development to in-vehicle implementation, Turing advances the entire stack in-house, with the goal of realizing the social implementation of fully autonomous driving, in which vehicles drive in place of humans under all conditions.

Company Overview

Company Name: Turing Inc. (Japanese: チューリング)
Headquarters: AE2-1-2, Tokyo Ryutsu Center Logistics Building A, 6-1-1 Heiwajima, Ota-ku, Tokyo, Japan
CEO: Issei Yamamoto
Founded: August 2021
Business: Development of fully autonomous driving systems
Website: https://tur.ing/

Careers

Turing is actively recruiting colleagues who will change the world by realizing fully autonomous driving from Japan. Please visit our careers page. We also regularly hold open offices, tech talks, and other events. For details, please see our Connpass page.

Media Contact

Turing Inc.
PR Representative (Abe): pr@turing-motors.com

Turing to Host “Turing AI Day 2025” on December 1Company to Present Reinforcement Learning and VLM-Based Autonomous Driving Models, and Outline Next-Generation GPU Infrastructure Plans

Turing Inc. (Headquartered in Shinagawa, Tokyo; CEO: Issei Yamamoto, hereinafter “Turing”) will host Turing AI Day 2025 on Monday, December 1, at Gate City Hall Ohsaki at 6:30 PM JST.

At the event, CEO Issei Yamamoto will present Turing’s upcoming strategies for both technological development and business expansion, followed by presentations from lead engineers showcasing the company’s latest research achievements and future development roadmap.

 Turing is a startup dedicated to the development of fully autonomous driving technology. The company is building an End-to-End (E2E) autonomous driving AI capable of handling perception, route planning, and vehicle control through a unified neural architecture. At the same time, Turing is developing large-scale foundation models that understand social norms, context, and common-sense reasoning. By integrating these two technologies, Turing aims to realize a future where vehicles can fully operate without human intervention, under all conditions.

 At this event, the company will share updates on the progress of its E2E autonomous driving AI, along with:

  • Turing’s next-generation GPU infrastructure concept.
  • new developments in autonomous driving models using reinforcement learning,
  • a Vision-Language Model (VLM)-based architecture for real-world driving, and

Program

  • Future Strategy for Technology and Business Development / CEO: Issei Yamamoto
  • Overview of AI Development / CTO: Yu Yamaguchi
  • E2E Autonomous Driving AI
  • Reinforcement Learning Model Development
  • VLA (Vision-Language-Action) Model Development
  • Next-Generation GPU Infrastructure
  • Q&A Session

Note: The agenda is subject to change.

Event Details

Title: Turing AI Day 2025
Date & Time: Monday, December 1, 2025, 6:30 PM – 8:15 PM (JST)
Format: Hybrid (in-person and online)
Venue: Gate City Hall Ohsaki (Gate City Ohsaki West Tower B1, 1-11-1 Osaki, Shinagawa, Tokyo)
Admission: Free (in-person attendance will be determined by lottery)
Registration: https://forms.gle/UHXGnDkKPRJonWXVA

Please register via the form by Thursday, November 27 at 11:00 PM JST.
Notifications regarding lottery results for in-person attendance and the online viewing URL will be sent to the email address provided by Friday, November 28.

About Turing

Company Name: Turing Inc. (Japanese: チューリング株式会社)
Headquarters: East Tower 4F, Gate City Ohsaki, 1-11-2 Osaki, Shinagawa-ku, Tokyo
Representative: Issei Yamamoto, CEO
Founded: August 2021
Business: Development of fully autonomous driving systems
Website: https://tur.ing/

Recruitment

Turing is actively hiring passionate individuals to join our mission to revolutionize the world through Japan-led fully autonomous driving technology. Please visit our recruitment page for more information.

We also regularly host open office events and technical talks. For event updates, please refer to our Connpass page.

Media Contact

PR Representative (Abe)pr@turing-motors.com

Turing Signs Contract with Japan’s Acquisition, Technology & Logistics Agency (ATLA) for E2E Autonomous Driving Verification Demonstrating the Versatility and Reliability of End-to-End Autonomous Driving AI in Off-Road Environments

Turing Inc. (Headquartered in Shinagawa, Tokyo; CEO: Issei Yamamoto; hereinafter “Turing”) is pleased to announce that they have signed a service contract with the Ground Systems Research Center of the Acquisition, Technology & Logistics Agency (ATLA) to verify its End-to-End (E2E) autonomous driving technology.

Verification Overview

 As part of the contract, Turing will collect baseline data in unpaved road environments and evaluate inference results generated by its E2E model under off-road conditions. The model’s output will control the vehicle and will evaluate the model’s behavior across a variety of terrain and surface types.

Going forward, Turing will continue to develop E2E autonomous driving systems for passenger vehicles with partners in the industrial and public sectors. These initiatives will allow Turing to contribute to safe, reliable mobility and ultimately achieve fully autonomous driving.

Contract Overview

Contract TitleVerification Services for End-to-End Autonomous Driving Technology
ClientGround Systems Research Center, Acquisition, Technology & Logistics Agency (ATLA)
Bidding MethodOpen Competitive Bidding
Implementation PeriodUntil March 23, 2026

About Turing

Turing is a Japan-based startup developing fully autonomous driving systems. Its End-to-End architecture allows AI to make all driving decisions based solely on camera input.

Turing’s innovations include:

Heron: a multimodal generative AI capable of high-level decision-making using various data inputs,

Terra: a generative world model that produces realistic video of driving scenarios,

CoVLA Dataset: a VLA model dataset that provides natural language-based descriptions and route planning from visual input.

Turing is making technological breakthroughs in autonomous driving to develop fully autonomous vehicles without steering wheels by 2030.

Company Overview

Company Name: Turing Inc.
Headquarters: East Tower 4F, Gate City Ohsaki, 1-11-2 Ohsaki, Shinagawa-ku, Tokyo
CEO: Issei Yamamoto
Founded: August 2021
Business: Development of fully autonomous driving technologies
Website: https://tur.ing/

Careers

Turing is hiring individuals passionate about transforming the world with fully autonomous Japanese driving technology. Please visit our careers page to learn more. We also host regular events such as open office sessions and Tech Talks.

Media Contact

PR Representative (Abe): pr@turing-motors.com

Turing Selected for Phase 3 of Japan’s GENIAC Program for Generative AI DevelopmentDeveloping an In-Vehicle Physical Foundation Model for Fully Autonomous Driving

Turing Inc. (Headquartered in Shinagawa, Tokyo; CEO: Issei Yamamoto, hereinafter “Turing”) is pleased to announce that they have been selected for Phase 3 of the “GENIAC” program, a national initiative led by Japan’s Ministry of Economy, Trade and Industry (METI) and the New Energy and Industrial Technology Development Organization (NEDO) to strengthen domestic generative AI development capabilities.

With this selection, Turing will receive government support for initiatives involving GPU-based computing resources and dataset construction. These efforts will accelerate the company’s development of an in-vehicle physical foundation model for fully autonomous driving.

This marks Turing’s third selection under the GENIAC program, following their inclusion in Phase 1 (announced in February 2024) and Phase 2 (announced in October 2024).

Phase 3 selection results can be found here: https://www.nedo.go.jp/koubo/CD3_100397.html

About the Project

In this project, Turing will develop a physical foundation model that operates in tandem with real vehicles, based on a multimodal AI capable of high-level understanding of text, video, sensor data, and control signals. The system will integrate a visual compression tokenizer and a 10-billion-parameter physical foundation model trained on large-scale datasets. These components will be distilled and compressed for seamless integration with in-vehicle systems.

During the development process, Turing leverages their proprietary visual data compression technology to convert images into sub-1KB tokens before entering them into the model. This will build a physical foundation model and supporting systems capable of unified inference and generation across vision, language, and action.

This approach allows for real-time, high-precision autonomous driving in standalone mode, even in rural or mountainous areas with unreliable network infrastructure. This will help resolve mobility challenges faced by elderly residents and people living in depopulated regions.

Looking ahead to the real-world deployment of fully autonomous driving systems, Turing plans to release portions of the foundational technologies developed through this project as reusable infrastructure applicable across multiple domains.

About GENIAC

The Generative AI Accelerator Challenge (GENIAC) is a national initiative jointly implemented by Japan’s Ministry of Economy, Trade and Industry (METI) and the New Energy and Industrial Technology Development Organization (NEDO) to strengthen Japan’s generative AI development capabilities. They provide compute resources to support the development of foundation models (the core technology behind generative AI) and offer assistance for data utilization and demonstration studies related to AI applications.
URL:https://www.meti.go.jp/policy/mono_info_service/geniac/index.html

About Turing

Turing is a Japan-based startup developing fully autonomous driving systems. Their End-to-End architecture allows AI to make all driving decisions based solely on camera input.

Turing’s innovations include:

  • Heron: a multimodal generative AI capable of high-level decision-making that can use multiple data types,
  • Terra: a generative world model that produces realistic driving scenarios as video,
  • CoVLA Dataset: a VLA model dataset that provides natural language-based understanding and route planning from visual inputs.

Company Overview

Company Name: Turing Inc.
Headquarters: East Tower 4F, Gate City Ohsaki, 1-11-2 Osaki, Shinagawa-ku, Tokyo
CEO: Issei Yamamoto
Founded: August 2021
Business: Development of fully autonomous driving technologies
URL: https://tur.ing/

Careers

Turing is hiring individuals passionate about transforming the world through fully autonomous driving technologies developed in Japan. Please visit our careers page to learn more. We also host regular events such as open office sessions and Tech Talks.

Upcoming Events

・Tuesday, July 15, 2025 | 7:00 PM – 8:00 PM (JST)
Turing Tech Talk #25
Design and Operation for Large-Scale Processing
https://turing.connpass.com/event/361118/

・Tuesday, July 22, 2025 | 7:00 PM – 8:00 PM (JST)
Turing Tech Talk #26
The Vision System Behind E2E Autonomous Driving: A Deep Dive into the Camera Processing Pipeline
https://turing.connpass.com/event/362593/

Media Contact

PR Representative (Abe): pr@turing-motors.com

World’s Top-Performing Japanese VLM at 15 Billion Parameters: Turing Releases “Heron-NVILA-Lite-15B” 2B Model Runs Fast on iPhone -Turing Also Releases part of the STRIDE-QA Dataset —One of the World’s Largest Autonomous Mobility Datasets-

Turing Inc. (Head Office: Shinagawa, Tokyo; CEO: Issei Yamamoto, hereinafter “Turing”) has released a new family of Japanese visual-language models: Heron-NVILA-Lite-15B/2B/1B. The 15B model, boasting 15 billion parameters, achieved a score of 73.5 on the Heron-Bench benchmark and surpassed other open-source models of similar scale across multiple Japanese visual-language evaluations. The 2B model, with 2 billion parameters, enables fully local and high-speed inference on iPhone devices.

This research was conducted as part of the Japanese Ministry of Economy, Trade and Industry (METI) and NEDO’s GENIAC (Generative AI Accelerator Challenge) program supporting generative AI research in Japan.
Turing has also released several major research assets, including:

  • MOMIJI, the world’s largest interleaved※1 Japanese-vision dataset.
  • STRIDE-QA, one of the world’s largest 3D datasets※2 for autonomous mobility, combining language and spatial-temporal information.

※1: A format that records data in a preserved sequence, allowing models to learn the connections between text, images, and the surrounding context.
※2: Based on Turing’s own research, this is the largest interleaved Japanese–image dataset known to date.

Background

Autonomous driving requires perception and decision-making systems that can instantly interpret real-world environments for safe operations. At the core of such systems are multimodal large language models (MLLMs), which acquire human-like common sense, background knowledge, and contextual understanding by learning from data such as images (vision), text (language), and embodied multimodal foundation models that build upon these MLLMs to learn from real-world sensor inputs to control outputs. However, high-quality Japanese-language training data that can simultaneously handle both vision and language is scarce, and there are few research cases involving lightweight high-performance MLLMs and embodied multimodal foundation models designed for in-vehicle implementation.

Turing has conducted multiple research and development initiatives as part of the GENIAC program, including the advancement of multimodal models, the construction of autonomous mobility datasets incorporating three-dimensional information, and the development of embodied autonomous driving models. The resulting model files and source code are now publicly available.

About Heron-NVILA-Lite-15B

Heron-NVILA-Lite-15B is a 15B-parameter open-source Japanese visual-language model designed to understand contextual and cultural background in Japanese. It scored 73.5 on the Heron-Bench benchmark for Japanese image-language tasks, outperforming other public models of similar scale (as of May 2025, per Turing’s evaluation).

This model uses an interleaved training format, where text and images are alternated during pre-training. This technique has proven effective in Japanese-language modeling.
 Model: Heron-NVILA-Lite-15B on Hugging Face
 Training details: Turing Tech Blog

About the Heron App for iOS

Turing also developed the Heron App for iOS, an image analysis AI app capable of fast local inference directly on smartphones. The app was optimized for offline performance on mobile devices by scaling the model down to 2 billion parameters

The app will be released on the App Store soon. Technical details and optimization strategies will be shared via Turing’s Tech Blog.

About the MOMIJI Dataset

MOMIJI (Modern Open Multimodal Japanese-filtered Dataset) is the world’s largest pre-training dataset for Japanese visual-language models in interleaved format. The dataset was released as 249 million image URLs paired with text in JSONL format. Further details will be provided in an upcoming post on our tech blog.

Hugging Face:https://huggingface.co/datasets/turing-motors/MOMIJI

About the STRIDE-QA Dataset

STRIDE-QA (SpatioTemporal Reasoning In Driving Environments QA) is one of the world’s largest 3D datasets for autonomous mobility. It was constructed by extracting 100 hours and 20,000 scenes from more than 3,500 hours of driving data collected by Turing in central Tokyo using cameras, LiDAR, and other sensors.

Each scene is annotated with consistent IDs and 3D bounding boxes for all traffic objects, such as vehicles and pedestrians. This allows for spatial tracking and continuous temporal tracking. A total of 12.63 million question–answer pairs have been generated, including both object-centric and ego-vehicle-centric questions such as “Is there a pedestrian on the crosswalk?” and “What will be the distance to the car ahead in two seconds?” These allow evaluation of an AI system’s ability to describe situations and predict future developments.

STRIDE-QA-mini, a partial version of the dataset that includes 200 scenes and approximately 120,000 Q&A pairs, has already been released to academic institutions. The full version is planned for a future release.
Hugging Face:https://huggingface.co/datasets/turing-motors/STRIDE-QA-Mini

The results presented in this press release are based on research supported by GENIAC (Generative AI Accelerator Challenge), a project led by Japan’s Ministry of Economy, Trade and Industry (METI) and the New Energy and Industrial Technology Development Organization (NEDO), aimed at strengthening domestic generative AI capabilities.
Reference press release:https://tur.ing/en/news/20241010

Company Overview

Company Name: Turing Inc.
Headquarters: East Tower 4th floor, Gate City Ohsaki, 1-11-2 Osaki, Shinagawa-ku, Tokyo
CEO: Issei Yamamoto
Founded: August 2021
Business: Development of fully autonomous driving technologies
URL: https://tur.ing/

Careers

Turing is hiring individuals passionate about transforming the world through fully autonomous Japanese driving technology. Please visit our careers page to learn more. We also host regular events such as open office sessions and Tech Talks.

Media Contact

PR Representative (Abe): pr@turing-motors.com

Turing Secures ¥1.325 Billion Loan from Mizuho Bank

Turing Inc. (Head Office: Shinagawa, Tokyo; CEO: Issei Yamamoto; hereinafter “Turing”) has entered into a loan agreement with Mizuho Bank, Ltd. (Head Office: Chiyoda, Tokyo; President & CEO: Masahiko Kato; hereinafter “Mizuho Bank”) for a total of ¥1.325 billion.

Turing is a startup dedicated to the development of fully autonomous driving technologies. The company is building an End-to-End (E2E) autonomous driving system in which AI makes all driving decisions—steering, acceleration, braking, and more—using only camera input.

 In addition to its E2E system, Turing is also developing a range of advanced technology that includes:

  • Heron, a multimodal generative AI that makes sophisticated decisions using multiple data sources;
  • Terra, a generative world model capable of producing realistic driving scene videos;
  • CoVLA Dataset, a VLA model dataset that enables detailed natural language descriptions of driving environments and optimal route planning based on image inputs.

Turing uses these innovations for technological advancement in autonomous driving. They will use the funds from this loan as working capital to support projects from national and local governments. Turing will continue to accelerate business development efforts toward achieving fully autonomous driving.

Company Overview

Company Name: Turing Inc.
Headquarters: Gate City Osaki East Tower 4th floor, 1-11-2 Osaki, Shinagawa-ku, Tokyo
CEO: Issei Yamamoto
Founded: August 2021
Business: Development of fully autonomous driving technologies
URL: https://tur.ing/

Careers

Turing is hiring individuals passionate about transforming the world through fully autonomous Japanese driving technology. Please visit our careers page to learn more. We also host regular events such as open office sessions and Tech Talks.

Media Contact

PR Representative (Abe):pr@turing-motors.com

Turing Selected for METI’s FY2024 Subsidy Program, Supporting the Development and Demonstration of Autonomous Driving Services that Address Regional Mobility Challenges

Turing Inc. (Head Office: Shinagawa, Tokyo; CEO: Issei Yamamoto; hereinafter “Turing”) is pleased to announce that it has been selected as a joint applicant for the Ministry of Economy, Trade and Industry (METI)’s FY2024 “Development and Demonstration Support for Autonomous Driving Services to Address Regional Mobility Challenges” supplementary subsidy program.

Through this program, Turing will accelerate the development of End-to-End autonomous driving systems while conducting real-world demonstrations to help solve mobility issues in underserved regions. The ultimate goal is to achieve the safe and widespread adoption of fully autonomous driving technology.

Selection announcement:https://www.meti.go.jp/information/publicoffer/saitaku/2025/s250319001.html

Turing’s Initiatives

Turing will undertake the following initiatives to construct a large-scale virtual dataset combining real-world driving data with multimodal generative AI, allowing for the development and validation of advanced autonomous driving technologies capable of handling complex traffic conditions.

Generation of Large-Scale Virtual Datasets

Enriching real-world data with spatiotemporal (4D) annotations—such as ego-vehicle viewpoints and the detailed movements of surrounding vehicles and pedestrians—will allow Turing to create realistic training datasets. These will be enhanced with natural language descriptions and data from domain-adapted generative AI, leading to the integrated learning of perception, prediction, and planning while accelerating the development of world models for autonomous driving.

Generative AI Model Development

The models will be trained using a combination of real-world data, simulation data, and world model outputs. Red-team* techniques will be employed to evaluate model vulnerabilities and generate adversarial scenarios, strengthening the robustness of End-to-End autonomous driving systems against perception errors and adversarial input.

Preparing for Dataset Public Releases

Turing will develop high-quality datasets that will be useful in building simulation platforms to speed up the safety and functionality testing of AD/ADAS systems. The datasets will be able to preemptively address the risks and concerns associated with generative AI technologies.

※Red-teaming: A method for assessing system security by simulating adversarial attacks to identify vulnerabilities.

About the FY2024 Supplementary Subsidy Program

Rural areas in Japan face growing mobility issues due to driver shortages and increased transportation demand, requiring new mobility solutions. METI’s mobility DX strategy calls for early commercialization of profitable autonomous taxi services in Japan, an area in which global competitors are already making significant progress.

The program promotes the creation of a Japanese standard model for autonomous driving centered on domestic automakers. Implementing this model across regions and eventually to similar global markets will enhance Japan’s industrial competitiveness.

To compete with U.S. and Chinese players in software-defined vehicle (SDV) development, a neutral entity will establish an “open dataset” using real-world and AI-generated virtual data. This will support the creation of realistic simulation environments and iterative safety validation cycles, ultimately accelerating advanced SDVs and autonomous driving software.

※Program details: https://tur.ing/wp-content/uploads/2025/11/k250205003_1.pdf

About Turing

Turing is a Japan-based startup developing fully autonomous driving systems. Its End-to-End architecture allows AI to make all driving decisions—from steering to acceleration and braking—based solely on camera input.

Turing’s innovations include:

  • Heron: a multimodal generative AI that makes sophisticated decisions using multiple data sources;
  • Terra: a generative world model capable of producing realistic driving scene videos;
  • CoVLA Dataset: a VLA model dataset that enables detailed natural language descriptions of driving environments and optimal route planning based on image inputs.

Turing is driving technological breakthroughs in autonomous driving to develop fully autonomous vehicles without steering wheels by 2030.

Company Overview

Company Name: Turing Inc.
Headquarters: Gate City Osaki East Tower 4th floor, 1-11-2 Osaki, Shinagawa-ku, Tokyo
CEO: Issei Yamamoto
Founded: August 2021
Business: Development of fully autonomous driving technologies
URL: https://tur.ing/

Careers

Turing is hiring individuals passionate about transforming the world through fully autonomous Japanese driving technology. Please visit our careers page to learn more. We also host regular events such as open office sessions and Tech Talks.

Media Contact

PR Representative (Abe):pr@turing-motors.com

Turing Secures a 500-Million Yen Loan Agreement with Sumitomo Mitsui Banking Corporation

Turing Inc. (Headquarters: Shinagawa, Tokyo; CEO: Issei Yamamoto; hereinafter referred to as “Turing”) has entered into a 500-million yen loan agreement with Sumitomo Mitsui Banking Corporation (hereinafter referred to as “SMBC”).

Turing is a startup dedicated to the development of fully autonomous driving technology. The company is developing an End-to-End (E2E) autonomous driving system in which AI makes all driving decisions using only data obtained from cameras.

Turing is also developing other cutting-edge technologies for autonomous driving. These include “Heron,” a multi-modal generative AI that makes sophisticated decisions using multiple data sources; “Terra,” a generative world model that can create realistic driving scenes as videos; and the “CoVLA Dataset,” a VLA model dataset that provides detailed natural language descriptions of driving environments for optimal route planning.

This loan will be used to develop autonomous driving systems and in joint research initiatives with partner companies, accelerating Turing’s business development toward achieving fully autonomous driving.

Company Overview

Company Name: Turing Inc.
Location: 4th floor, East Tower, Gate City Osaki,1-11-2 Osaki, Shinagawa, Tokyo, Japan
Representative CEO Issei Yamamoto
Founded: August 2021
Business: Development of fully autonomous driving technology
URL: https://tur.ing/

Career Opportunities

Turing is seeking passionate individuals to join our team and help us change the world through fully autonomous driving technology from Japan. Please visit our recruitment page for details. We also regularly host open office events and tech talks. For more information, please check our Connpass page.

Media Contact 

PR Representative
(Abe): pr@turing-motors.com

Turing Develops Visual Data Compression Technology for Large-Scale AIAchieving Significant Capacity Reduction with High-Quality Reconstruction

Turing Inc. (Headquartered in Shinagawa, Tokyo; CEO: Issei Yamamoto, hereinafter “Turing”) has developed a new technology (patent pending) that efficiently compresses large-scale video and image data while retaining high accuracy in a format optimized for AI. By combining a learning mechanism that aggregates key information locally with a data allocation approach based on importance, this technology enables high-speed and high-precision data utilization for autonomous driving AI, multimodal AI, and more.

Release Background

In recent years, multimodal large language models (MLLMs), which handle multiple data types such as images and text simultaneously, have drawn increasing attention. The demand for advanced development capable of processing vast amounts of data continues to rise. However, conventional image embedding techniques face challenges in efficiently delivering information in a format truly optimized for AI.

Technology Overview

Turing’s newly developed technology provides a mechanism to compress massive amounts of data while preserving essential information with high accuracy. By converting various data—such as text and images—into a sequence of tokens (the smallest units processed by AI) and introducing a variable-length compression approach, users can add or remove tokens as needed. This makes it possible to significantly reduce data size without compromising required image quality or analytical precision.

A technique called Tail Token Drop is employed during training. It randomly deletes tokens from the end of the sequence and compares the differences to optimize the model. This process effectively concentrates critical information toward the beginning of the token sequence, minimizing the loss of crucial data even under higher compression ratios.

Furthermore, images can be reconstructed from the token sequences. Compared to traditional image formats such as JPEG or WebP, this method can recreate visually natural images using fewer bytes, opening up new possibilities for real-time autonomous driving systems and cloud-based platforms where communication costs and responsiveness are vital.

Furthermore, details on this technology have been published in the paper One-D-Piece: Image Tokenizer Meets Quality-Controllable Compression and are explained in depth on our tech blog. The model files and source code are available for commercial use under the Apache License 2.0.

Project Pagehttps://turingmotors.github.io/one-d-piece-tokenizer/
Tech Bloghttps://zenn.dev/turing_motors/articles/6d77c5a3b3712e
s-1460x1142_v-fms_webp_aa827448-6fba-4a43-984b-040cc1c44ac5_small

Potential Applications

1. Autonomous Driving

By compressing onboard camera footage into fewer tokens, this technology can deliver visual data to autonomous driving base models more efficiently. Preventing excessive computational load when handling large volumes of data enables large-scale AI models to rapidly recognize and evaluate the surrounding environment in real time.

2. Multimodal Models and World Models

Images and videos tokenized using this technology are expected to be directly used as inputs and outputs for MLLMs or world models, similar to language tokens. By adjusting the number of tokens according to context and data volume, one can ease the computational burden of training and inference in these models while maintaining overall accuracy.

Company Overview

Company Name: Turing Inc.
Location: 4th Floor, East Tower, Gate City Osaki, 1-11-2 Osaki, Shinagawa-ku, Tokyo
Representative: CEO Issei Yamamoto
Founded: August 2021
Business: Development of fully autonomous driving technology
URL: https://tur.ing/

Career Opportunities

Turing is seeking individuals who are eager to change the world by making fully autonomous driving a reality. We frequently host company introduction events and autonomous driving experience sessions, so feel free to reach out to us.
Careers Page: https://tur.ing/jobs
Event information: Connpass

Media Inquiries

PR Contact (Hiraku Abe): pr@turing-motors.com

Turing Secures ¥1 Billion Funding from Government-Backed Fund ‘JIC VGI,’ etc.,Raising a Total of ¥5.5 Billion in Pre-Series A Round

Turing Inc. (Shinagawa, Tokyo; CEO: Issei Yamamoto, hereafter referred to as “Turing”) has announced that they have secured ¥1.02 billion at the completion of an additional Pre-Series A funding round. This includes ¥820 million in equity financing from JIC Venture Growth Investments, JR West Innovations, NCB Venture Capital, Turing CFO Masato Morishima, and a ¥200 million loan from Mizuho Bank. The total amount raised in the Pre-Series A round has reached ¥5.558 billion.。

Turing is a startup dedicated to the development of fully autonomous driving technology. The company is creating an End-to-End (E2E) autonomous driving system in which AI makes all the decisions—such as steering, acceleration, and braking—using data obtained exclusively from cameras.

Turing drives innovation in the autonomous driving field by developing advanced technologies such as the multimodal generative AI “Heron,” which makes sophisticated decisions using multiple types of data; the generative world model “Terra,” capable of creating realistic driving scenario videos for autonomous driving; and the “CoVLA Dataset,” a VLA model dataset for autonomous driving that provides detailed natural language descriptions of driving environments and generates optimal route plans. The company’s goal is to develop fully autonomous vehicles without steering wheels by 2030. 

The funds raised in this round will be used to execute the “Tokyo30” project, which will develop an autonomous driving system capable of operating in Tokyo for 30 minutes without human intervention by 2025. The funds will also support the recruitment of talent, particularly machine learning engineers, to accelerate these initiatives.

Comments from Investors (in Alphabetical Order)

Toshiya Kishimura, Venture Capitalist at JIC Venture Growth Investments Co., Ltd.

AI technology is moving into a new stage of real-world interaction, following innovations in computer vision and natural language processing. Turing’s challenge to achieve fully autonomous driving through E2E (End-to-End) approach stands at the forefront of this advancement. Their proprietary technologies, including the multimodal generative AI ‘Heron’ and the generative world model ‘Terra,’ hold the potential to bring us closer to fully autonomous driving – one of humanity’s grand challenges. We highly value Turing’s innovative approach and technical capabilities, and will support their challenge toward realizing next-generation mobility. We expect this to bring about a new paradigm shift in the mobility industry.

Ryo Kawamoto, President of JR West Innovations

Autonomous driving is a challenging endeavor, and I am confident that Turing, with its outstanding leadership team and talented engineers, has the capability to achieve it. JR West Group is honored to be part of this groundbreaking project, and we believe that this challenge will bring new value to local communities and to the mobility experience. Together with Turing, we are committed to creating an exciting future.

Hiroki Hayashi, President of NCB Venture Capital Co., Ltd.

We are honored to have the opportunity to invest in Turing, a Japanese company that leads the way in fully autonomous driving. Turing is at the forefront of the domestic development of End-to-End (E2E) fully autonomous driving AI and has established itself as a top-tier player globally. We were impressed by the company’s vision, technological expertise, and the passion of its talented leadership team, which led us to make this investment. We look forward to seeing Turing contribute to Japan’s world-renowned automotive industry and make a significant impact on the growth of the domestic economy.

Recruitment Event Announcement

We will be holding a recruitment event at Turing’s headquarters on January 15 (Wed) and January 16 (Thu). Our executive team and engineers will be present to share insights on our development approach, strategies, and ongoing projects. The event will also feature team and project introductions, roundtable discussions, and networking opportunities. We invite you to join us and look forward to meeting you!

Event Name: Engineer Recruitment Online Information Session
Date and Time: January 15 (Wed), 18:00 – 18:45
Location: Online (No participant limit)
Target Audience: Mid-career engineers and engineering students interested in joining Turing
For more details: https://turing.connpass.com/event/336615/

Event Name: Open Office
Date and Time: January 16 (Thu), 18:30 – 20:00
Location: Turing Headquarters Office
Address: 1-11-2 Osaki, Shinagawa-ku, Tokyo, Gate City Osaki East Tower, 4th Floor
Number of Participants: 12 people
For more details: https://turing.connpass.com/event/339230/

About Turing

Company Name: Turing Inc.
Location: 1-11-2 Osaki, Shinagawa-ku, Tokyo, Gate City Osaki East Tower, 4th Floor
CEO: Issei Yamamoto
Founded: August 2021
Business: Development of fully autonomous driving systems
URL: https://tur.ing/

Career Information

We are seeking individuals who want to change the world by achieving fully autonomous driving technology in Japan. We frequently hold company introduction events and autonomous driving experience sessions, so feel free to contact us for more information.
Recruitment Pagehttps://tur.ing/en/jobs

Media Contact

PR Contact (Hiraku Abe): pr@turing-motors.com