iGEN
Visit IGEN World Explore IGEN Expo
EXPLORE UPGRADE PLANS
BREAKING
Tin Prices Poised to Rule Elevated in 2026 on Semiconductor Demand and Supply Disruptions India must boost oilseed yields to cut edible oil imports, SEA chief says India Air Freights 5 Tonnes of Medical Aid to Afghanistan Under Humanitarian Assistance Tsakos Joins Greek Capesize Ordering Wave at Hengli Heavy Industries How US quietly kept Gulf crude moving despite Iran's Hormuz blockade Rupee Rebounds 31 Paise to 94.29 as Easing Oil, Dollar Index Boost Sentiment Shipping Braces for Monster El Niño as NOAA Warns of Record-Intensity Event Threatening Global Trade Lanes India May Require Refiners to Triple Crude Oil Inventories After Lessons From China Fleets Reposition for Hormuz Reopening Ahead of US-Iran Peace Deal Signing Gold price prediction today: Central bank buying, US-Iran peace deal support gold above $4,300/oz Tin Prices Poised to Rule Elevated in 2026 on Semiconductor Demand and Supply Disruptions India must boost oilseed yields to cut edible oil imports, SEA chief says India Air Freights 5 Tonnes of Medical Aid to Afghanistan Under Humanitarian Assistance Tsakos Joins Greek Capesize Ordering Wave at Hengli Heavy Industries How US quietly kept Gulf crude moving despite Iran's Hormuz blockade Rupee Rebounds 31 Paise to 94.29 as Easing Oil, Dollar Index Boost Sentiment Shipping Braces for Monster El Niño as NOAA Warns of Record-Intensity Event Threatening Global Trade Lanes India May Require Refiners to Triple Crude Oil Inventories After Lessons From China Fleets Reposition for Hormuz Reopening Ahead of US-Iran Peace Deal Signing Gold price prediction today: Central bank buying, US-Iran peace deal support gold above $4,300/oz
Home ›› Technology ›› Ai ›› Decart's Oasis 3: Photorealistic Driving Simulations

Decart's Oasis 3: Photorealistic Driving Simulations

Decart has launched Oasis 3, a world model that simulates photorealistic driving environments in real time. Targeting autonomous vehicle companies, the model is available via API and aims to build a developer ecosystem. Despite its efficiency and photorealism, the model faces challenges in maintaining thematic consistency and physics simulation.

iG
iGEN Editorial
June 10, 2026
Decart's Oasis 3: Photorealistic Driving Simulations

AI startup Decart has introduced Oasis 3, an interactive world model designed to generate photorealistic driving environments in real time, according to TechCrunch. This model is primarily aimed at autonomous vehicle companies needing to simulate rare driving scenarios at scale. Decart plans to expand its use to robotics and other physical AI applications.

Developer Ecosystem and Pricing

Decart is offering Oasis 3 via API, hoping to cultivate a developer community similar to what OpenAI achieved with language models. Dean Leitersdorf, co-founder and CEO of Decart, envisions a robust developer ecosystem emerging around this technology. The model is priced at $0.02 per second, with enterprise pricing varying by use case.

Competitive Landscape

The world model arena is becoming increasingly competitive. Last year, Google released Genie 3 in research preview, while World Labs launched Marble for commercial use. Other companies like Luma and Runway are also adapting their video models into world models. Decart differentiates itself with the photorealism and efficiency of its models, powered by the Decart Optimization Stack (DOS), which allows models to run efficiently on Nvidia, Amazon, and Google hardware.

Technical Challenges

Despite its advantages, Oasis 3 faces challenges in maintaining thematic consistency and simulating physics accurately. The model is auto-regressive, generating one frame at a time and looking back at previous frames to decide the next. This method is compute-intensive and can lead to inconsistencies, such as environments changing unexpectedly and cars driving through each other.

Future Developments

To address these issues, Decart is working on improving the model's memory length and exploring ways to compress memory into fewer tokens. Leitersdorf anticipates that future versions will allow users to generate worlds based on video rather than images, potentially solving some of the current limitations.

"It takes me back to the early days of LLMs, when OpenAI invented the API for models," said Leitersdorf, highlighting the potential for developers to create new applications with Oasis 3.

Funding and Strategic Partnerships

Recently, Decart raised $300 million, increasing its valuation to nearly $4 billion. Strategic investors include Toyota, Adobe, and eBay, all potential customers. Nvidia also participated in the funding round, underscoring the strategic importance of Decart's technology.

Company Product Feature
Decart Oasis 3 Photorealistic driving simulations
Google Genie 3 Research preview world model
World Labs Marble Commercial use world model

Decart's Oasis 3 represents a significant step forward in the simulation of driving environments, offering potential cost savings and efficiency improvements for autonomous vehicle developers. However, the model's current limitations highlight the ongoing challenges in the field of world models.


Sources: TechCruunch

Keep Reading

Recommended Stories

Enterprise AI Takes Center Stage at VivaTech 2026 Technology

Enterprise AI Takes Center Stage at VivaTech 2026

VivaTech 2026 will spotlight enterprise AI, focusing on its integration into complex systems like logistics and manufacturing. The event will address challenges such as governance and compliance in AI deployment.

June 10, 2026
Stop treating AI as the strategy — focus on business outcomes instead Technology

Stop treating AI as the strategy — focus on business outcomes instead

TechRadar cautions against treating AI as a strategy itself, warning that organizations risk losing sight of business outcomes. The article emphasizes that AI should be a tool within a broader transformation strategy, requiring collaboration across departments, clear problem identification, and measurement beyond productivity gains.

June 16, 2026
LLM Jaggedness Unlocks Scientific Creativity: New Benchmark Reveals Uneven AI Capabilities Can Be Harnessed for Innovation Technology

LLM Jaggedness Unlocks Scientific Creativity: New Benchmark Reveals Uneven AI Capabilities Can Be Harnessed for Innovation

A new arXiv paper introduces SciAidanBench, a benchmark for measuring the scientific creativity of large language models. The research finds that LLM capabilities are jagged—uneven across tasks and domains—but that this jaggedness can be harnessed through ensemble methods to produce superior scientific ideas.

June 16, 2026
First Model-Free Universal AI Agent Proved Asymptotically Optimal in General Reinforcement Learning Technology

First Model-Free Universal AI Agent Proved Asymptotically Optimal in General Reinforcement Learning

Researchers introduced Universal AI with Q-Induction (AIQI), the first model-free agent proven asymptotically ε-optimal in general reinforcement learning. Unlike previous model-based optimal agents like AIXI, AIQI performs induction over action-value functions. The proof also establishes optimality for Self-AIXI without ad-hoc assumptions.

June 16, 2026