Chinese media tested ADAS and FSD in various scenarios, including highways and night driving. Tesla vision-based system significantly outperformed emerging Chinese brands like Huawei & Xiaomi, as well as traditional manufacturers. Even with LiDAR, competitors’ ADAS performance lags behind Tesla. Only 6 out of 34 other systems passed half of the tests (3 out of 6) and 24 passed one or none of the tests. Tesla had vehicles in the tests and Tesla had top performance by passing 5 out of 6 tests.
The test scenarios were
Sudden Steering by Lead Vehicle – Highway Scenario
Temporary Construction Zone
Construction Zone with Stationary Truck
Crashed Vehicle Spotted on Highway
Aggressive Merge at On-Ramp
Boar Darting Across Highway
The construction zone with stationary truck, boar (wild pig) darting across highway and the aggressive merge at on ramp were problems for almost all of the systems. About half of China systems could pass one or none of the tests.

NEWS: Chinese media tested ADAS in various scenarios, including highways & night driving. @Tesla’s vision-based system significantly outperformed emerging Chinese brands like Huawei & Xiaomi, as well as traditional manufacturers. Even with LiDAR, competitors’ ADAS performance… pic.twitter.com/xW5CmmbnQ4
— Sawyer Merritt (@SawyerMerritt) July 23, 2025
Rankings
Top Performers:
• Tesla Model 3 (2023 RWD Refreshed) – 5/6
• Tesla Model Y (2023 LR Refreshed) – 5/6
• Avita 12 (2025 Dual Motor Max) – 3/6
• Xiaomi SU7 (2025 Max) – 3/6
• IM LS6 (2025 625km Long Range Max) – 3/6
• Voyah FREE (2025 Ultra 6C Edition) – 3/6
• Zhiji LS7 (2024 Dual Motor 100kWh) – 3/6
• Leapmotor C11 (2025 YOU Edition 100kWh 4WD) – 3/6
Mid-Tier:
• Aion RT (2025) – 2/6
• Geely Galaxy E8 (2025 520 Pro Light) – 2/6
• AITO M5 (2024 650 Smart Drive Pro) – 2/6
• Xpeng G6 (2024 Ultra Rear Drive) – 2/5
• Nio ET5 (2024 75kWh) – 2/5
Low Performers:
• Li Auto L6 (2025 Max Smart Driving) – 1/6
• Xiaomi SU7 (2025 Standard Rear Drive Max) – 1/5
• Voyah Dreamer (2025 Ultra 6C Edition) – 1/5
• BYD Song L DM-i (2025 DM-i 80KM) – 1/5
• ICAR V23 (2025 501km Rear Drive) – 1/5
• Xiaomi SU7 Ultra (2025) – 1/4
• Nezha S (2025 305KM Edition) – 1/4
• BYD Seal 06 (2025 305KM Edition) – 1/4
• Rising Auto F7 (2024 100kWh) – 1/4
Failed All Tests (0 Passes):
• Zeekr 001 (2025 YOU Edition 100kWh 4WD) – 0/6
• BMW 3 Series PHEV (2025 Long Range) – 0/6
• Cadillac CT6 (2024 900) – 0/6
• BYD Han EV (2025 701KM) – 0/5
• Nio EC6 (2024 530KM Light Sensing) – 0/5
• Changan Deepal SL03 (2025 Pro 730STI Dragon Edition) – 0/5
• GAC Aion Hyper GT (2025 Long Range Pro) – 0/5
• BYD Han DM-i (2025 125KM Light Sensing) – 0/5
• Zeekr 7X (2025 100kWh) – 0/5
• Xiaomi SU7 (2024 Light Sensing Max) – 0/4
• BYD Qin DM-i (2025 115KM) – 0/4
• Leapmotor C01 (2025 525KM High Speed) – 0/4
• Chery Arrizo 8 (2025 260T) – 0/4
• Jikrypton Star Era (2025 Max Drive Pro Edition) – 0/4
• firefly (2025 Lite) – 0/4

Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
Tesla’s remains a publicity stunt. Opensource data show that they are not even getting 200 miles between critical (likely accident avoiding) disengagements in cities with latest version of FSD software. That is no better than they were doing with some versions 1.5 years ago so development has clearly stalled out at about 0.1% of critical disengagement reliability they need to find acceptance (hundreds of thousands of miles between critical disengagements).
https://app.powerbi.com/view?r=eyJrIjoiNTAzODg0MDgtYzI1Zi00MzBmLWEwZTAtOWI5OTAyNmU2ZGM0IiwidCI6ImMxM2M0MmQ1LTlhNTAtNDY3YS05Yjc3LWI1MjkyYzgxZjM1NSIsImMiOjF9
In other words they need to eliminate at least 999 out of every 1000 critical screwups currently being made, and there is zero hope of that happening in next year or two. They’ll need at minimum a huge increase in onboard hardware capabilities followed by a massive amount of training and checking.
So it will remain a loss-making human supervised stunt for the foreseeable future, not a realistic market player.