LLM.co Study Reveals Hybrid AI Strategies Driving Open Source LLM Adoption – Open Source For You
A new LLM.co study shows enterprises expanding open source LLM adoption to reduce costs, improve control, and build flexible hybrid AI infrastructure.
A new study from LLM.co indicates that enterprises are increasingly turning to open-source large language models (LLMs) as artificial intelligence shifts from experimentation to long-term infrastructure strategy.
The report finds that organisations are prioritising open-source models for flexibility, cost control, and deployment autonomy as AI becomes embedded in business operations. Today, 78% of organisations use AI in at least one business function, while generative AI adoption has reached 71%, signalling that LLMs are evolving into core enterprise tools rather than experimental technologies.
Research cited from McKinsey & Company shows that more than half of enterprises already use open-source AI tools somewhere in their technology stack, particularly in data science workflows, model development environments, and internal AI experimentation. Lower licensing costs, customisation flexibility, and the ability to run models on private infrastructure are key drivers of adoption.
However, proprietary platforms continue to dominate production environments. Mid-2025 data shows closed-source LLMs still account for roughly 87% of deployed enterprise workloads, particularly in high-reliability and customer-facing applications.
“Enterprises are making pragmatic decisions,” said Samuel Edwards. “Open source LLMs are gaining traction because they offer flexibility, control, and economic advantages. But closed models still deliver faster deployment and proven performance for many mission-critical applications. The future isn’t open versus closed — it’s strategic use of both.”
Enterprise sentiment data suggests the balance may shift. Forty-one percent of organisations plan to expand open-source LLM usage, while another 41% say they would transition from proprietary models once performance parity is reached, potentially leading to a more balanced ecosystem in the coming years.
Industry observers also note the rise of hybrid AI architectures. According to research cited from Forbes, 37% of enterprises now advocate hybrid AI stacks combining open-source and proprietary models to optimise cost, performance, governance, and vendor risk.
Open-source models are particularly attractive for on-premises deployment, stronger data governance, and reduced vendor lock-in—factors that are increasingly important in regulated and data-sensitive sectors. Proprietary platforms, however, continue to offer advantages in faster deployment, enterprise support, and shorter time-to-value for non-technical teams.
“Enterprises are no longer testing LLMs — they’re operationalizing them,” said the Chief Revenue Officer at LLM.co. “That means model decisions now affect cost predictability, compliance posture, infrastructure strategy, and total cost of ownership. The most sophisticated organizations are designing AI frameworks that capture the strengths of both ecosystems rather than committing to one camp.”
The report concludes that enterprise LLM adoption is entering a new phase, shifting from experimentation to optimisation as organisations prioritise flexibility, portability, financial efficiency, and infrastructure control in their AI strategies.
source
This is a newsfeed from leading technology publications. No additional editorial review has been performed before posting.
