TLDR: Stockholm-based AI research lab Farang has successfully raised €1.5 million in seed funding to develop and launch its next-generation foundational Large Language Models (LLMs). The funding round was led by Voima Ventures and the Amadeus APEX Technology Fund, with additional support from notable angel investors. Farang’s innovative architecture aims to outperform existing Transformer-based models like ChatGPT, Claude, and Gemini, particularly in specialized domains, while significantly reducing computational resource requirements.
Stockholm, Sweden – August 27, 2025 – Farang, a pioneering Swedish AI research lab dedicated to developing next-generation foundational Large Language Models (LLMs), today announced it has successfully secured €1.5 million in seed funding. This investment is earmarked to scale its proof-of-concept models and bolster its computational power for training and fine-tuning models tailored for specific specializations. The funding round saw leadership from Voima Ventures and the Amadeus APEX Technology Fund, complemented by contributions from prominent entrepreneur angel investors including Tero Ojanpera (Co-founder, Silo AI), Nilay Oza, and Niraj Aswani (Former Founders, Klevu). The company’s post-money valuation stands at €10 million.
Farang distinguishes itself by developing a completely new foundational architecture for LLMs, a direct competitor to the widely adopted Transformer architecture that underpins current market leaders such as ChatGPT, Claude, and Gemini. Emil Romanus, Founder of Farang, emphasized their unique approach: “We’re not building another application layer on top of existing models. We’ve developed a completely new foundational architecture that enables us to create specialised AI assistants that outperform current solutions in specific domains like programming and medicine, while using twenty-five times less computational resources. Based on current testing, we believe the percentage of resources used will decrease even further in the future.”
Founded in 2025, Farang’s innovative architecture operates by comprehending the complete response first – akin to a human imagining a picture before painting it – and then translating that concept into words. This contrasts with traditional LLMs that primarily rely on word-by-word prediction. While contemporary AI models have incorporated reasoning steps to simulate ‘thinking time,’ they still fundamentally depend on word prediction. Farang’s method conducts this internal reasoning through non-textual mechanisms, resulting in more coherent and contextually rich responses while drastically cutting down computational demands.
The company, comprising a lean team of five researchers, developers, and business professionals, is spearheaded by Emil Romanus, an engineer boasting over two decades of experience in search, AI, and previous startup ventures. Farang plans to initially target niche applications where existing AI assistants often struggle or underperform. These include specialized AI assistants for particular programming languages and frameworks, specific medical fields, and internal company AI tools. Their immediate focus is on React programming, with the goal of generating more optimized and iterable code than current LLMs. Furthermore, Farang’s technology offers organizations the capability to deploy specialized AI models on-premises, ensuring full privacy controls – a significant advantage for sectors handling sensitive data like healthcare, legal, and financial services.
Also Read:
- Nugen Secures Pre-Seed Funding to Advance Reliable AI for Enterprise Applications
- Holiwise Secures €1.45 Million Pre-Seed Funding to Advance AI-Powered Premium Travel Platform
Looking ahead, Farang has ambitious plans to become a market leader in artificial intelligence, with the ultimate aim of competing with and surpassing OpenAI in the general AI space. Inka Mero, Managing Partner & Founder of Voima Ventures, expressed confidence in Farang’s potential: “We look for exceptional Founders and technologies that reset the curve – not just optimise around the edges. Farang showcases how Europe can step up in the global AI race, as its foundational architecture provides a true paradigm shift – specialised, efficient, and enterprise-grade from day one.”
