London AI Lab Inherent Claims Research Breakthrough with Smaller Model Outperforming Claude and GPT-5.5
DeepMind alumni startup Inherent says its Faraday agent outperformed Anthropic and OpenAI models at replicating scientific research findings.
Inherent’s Faraday Agent Outperforms Larger Competitors
Inherent, a London-based AI lab founded by Google DeepMind alumni, claims its Faraday agent has outperformed larger models from Anthropic and OpenAI at independently reproducing the findings of published scientific papers without being told the answer in advance.
The agent was benchmarked against Anthropic’s Claude Opus 4.8 and OpenAI’s GPT-5.5, both described as frontier-scale systems. Faraday runs on Qwen 3.6 with 27 billion parameters—substantially smaller than the competing models.
How Inherent Built Faraday
Inherent used reinforcement learning as its primary training method for Faraday, rewarding the agent for good outcomes rather than specifying explicit rules. Notably, Faraday uses OpenAI’s GPT-5.5 Codex for coding tasks rather than a proprietary coding tool built by Inherent.
Funding and Team
Inherent raised a $50 million seed round before releasing Faraday. The lab was co-founded by Edward Hughes, Louis Kirsch, Kaloyan Aleksiev, and Tantum Collins, and is based in King’s Cross, London. The company currently has approximately 12 employees and plans to grow headcount to about 20 to 25 by end of 2026.
Source: TechCrunch