CREMER's FermBase turns its Differential Bio collaboration into a strategic partnership
Hamburg trading group CREMER formalises its fermentation media partnership with Munich spin-off Differential Bio after 3,300 automated experiments.
FermBase, the fermentation media brand of the Hamburg trading group CREMER, has turned its working collaboration with the Munich startup Differential Bio into a formal strategic partnership. Differential Bio dated the agreement 15 October 2026 on its own news page, and the trade press reported it on 16 October. The two companies say they will develop further products and services for fermentation companies, covering the span from a first experiment through to industrial production. CREMER contributes raw materials and supply chains, Differential Bio contributes robotic laboratory automation and machine learning. Financial terms, contract duration and any equity component were not disclosed.
The partnership follows a product rather than preceding one. CREMER launched FermBase in February 2026 to supply fermentation companies with carbon and nitrogen sources, mineral salts, trace elements, custom formulations and standardised blends, from pilot scale through to industrial production. Its first product, Y-Base Blend, is a standardised trace element blend for yeast fermentation in defined media. Differential Bio screened the problem down from more than 50,000 possible trace element combinations to 3,300 automated experiments across different yeast strains. Nine high performing formulations went forward to microlitre scale testing, and one was selected for bioreactor use. The companies published the work as a joint whitepaper on 1 October 2026 and put the development time at six months.
The trial data is what makes the blend sellable. In an early access programme with selected partners across different processes and products, the blend performed comparably in every case, and according to Cremer it beat customers' own in-house recipes in a large share of them. Gerlis von Haugwitz, product manager at Cremer, has said the point of the programme was to validate the blend inside customer bioreactor processes rather than only under laboratory conditions. Y-Base Blend ships in 1 kg packs for laboratory trials, 25 kg for pilot and scale-up work and multi tonne quantities for industrial production, with composition and specification held identical across all three sizes.
Differential Bio is a spin-off from the Technical University of Munich. It came out of stealth on 26 March 2025 with a €2 million pre-seed round led by Ananda Impact Ventures and ReGen Ventures, with Carbon13, Climate Capital, Better Ventures and CDTM Ventures also taking part alongside a group of angels. The founding team is Christian Spier as chief executive, Martin Patz as chief technology officer and Dong Zhao as chief scientific officer. Its Virtual Scale-up Platform combines miniaturised fermentation, robotic laboratory automation and AI models that simulate and optimise a bioprocess before anyone pays for a physical scale-up run. At the pre-seed the company reported a fourfold increase in biomass yield and a 16 per cent reduction in production costs in a probiotics case. Christian Flach, chief executive of CREMER, framed the deal as the group rethinking its existing business, while Spier framed it as moving the shift to bio-based production forward.
Why it matters for AgTech founders: the instructive part is who paid, and with what. A commodity trader founded in 1946 did not build a data science team and did not write a cheque into a cap table. It bought the capability with raw material supply and distribution, which is currency it already had. For a deep tech company selling process optimisation into biomanufacturing, a trading group is a better first customer than a fermentation scale-up: it has the supply chain, the balance sheet and a commercial reason to want a differentiated product, where a cash constrained scale-up has none of the three. Note also what the model work turned into. It was not a software subscription. It was somebody else's SKU, sold by the bag, with the modelling embedded invisibly in the formulation. That route converts model output into revenue without having to sell software to a conservative buyer, and more AgTech AI companies should price it seriously. Third, media and raw material cost is one of the quietly decisive lines in any cost per kilogram model for a fermented ingredient, and that input chain is now being standardised by people who trade commodities for a living. Lower cost and less batch variability are good for the sector, but a founder whose in-house media recipe was on the differentiation slide should take it off. Finally, look at the sequence and the clock: pre-seed in March 2025, a customer case study, a co-developed product, a joint whitepaper, then a named partnership with a group of this size, all inside about eighteen months. Each step handed the corporate something it could show internally, which is how a small company gets a large one to sign.
CREMER's FermBase turns its Differential Bio collaboration into a strategic partnership
A family owned raw materials trader founded in 1946 is buying an AI and robotics capability with supply chain rather than equity. The first joint product was a bag of trace elements.
FermBase, the fermentation media brand of the Hamburg trading group CREMER, has turned its working collaboration with the Munich startup Differential Bio into a formal strategic partnership. Differential Bio dated the agreement 15 October 2026 on its own news page, and the trade press reported it on 16 October. The two companies say they will develop further products and services for fermentation companies, covering the span from a first experiment through to industrial production. CREMER contributes raw materials and supply chains, Differential Bio contributes robotic laboratory automation and machine learning. Financial terms, contract duration and any equity component were not disclosed.
The partnership follows a product rather than preceding one. CREMER launched FermBase in February 2026 to supply fermentation companies with carbon and nitrogen sources, mineral salts, trace elements, custom formulations and standardised blends, from pilot scale through to industrial production. Its first product, Y-Base Blend, is a standardised trace element blend for yeast fermentation in defined media. Differential Bio screened the problem down from more than 50,000 possible trace element combinations to 3,300 automated experiments across different yeast strains. Nine high performing formulations went forward to microlitre scale testing, and one was selected for bioreactor use. The companies published the work as a joint whitepaper on 1 October 2026 and put the development time at six months.
The trial data is what makes the blend sellable. In an early access programme with selected partners across different processes and products, the blend performed comparably in every case, and according to Cremer it beat customers' own in-house recipes in a large share of them. Gerlis von Haugwitz, product manager at Cremer, has said the point of the programme was to validate the blend inside customer bioreactor processes rather than only under laboratory conditions. Y-Base Blend ships in 1 kg packs for laboratory trials, 25 kg for pilot and scale-up work and multi tonne quantities for industrial production, with composition and specification held identical across all three sizes.
Differential Bio is a spin-off from the Technical University of Munich. It came out of stealth on 26 March 2025 with a €2 million pre-seed round led by Ananda Impact Ventures and ReGen Ventures, with Carbon13, Climate Capital, Better Ventures and CDTM Ventures also taking part alongside a group of angels. The founding team is Christian Spier as chief executive, Martin Patz as chief technology officer and Dong Zhao as chief scientific officer. Its Virtual Scale-up Platform combines miniaturised fermentation, robotic laboratory automation and AI models that simulate and optimise a bioprocess before anyone pays for a physical scale-up run. At the pre-seed the company reported a fourfold increase in biomass yield and a 16 per cent reduction in production costs in a probiotics case. Christian Flach, chief executive of CREMER, framed the deal as the group rethinking its existing business, while Spier framed it as moving the shift to bio-based production forward.
Why it matters for AgTech founders: the instructive part is who paid, and with what. A commodity trader founded in 1946 did not build a data science team and did not write a cheque into a cap table. It bought the capability with raw material supply and distribution, which is currency it already had. For a deep tech company selling process optimisation into biomanufacturing, a trading group is a better first customer than a fermentation scale-up: it has the supply chain, the balance sheet and a commercial reason to want a differentiated product, where a cash constrained scale-up has none of the three. Note also what the model work turned into. It was not a software subscription. It was somebody else's SKU, sold by the bag, with the modelling embedded invisibly in the formulation. That route converts model output into revenue without having to sell software to a conservative buyer, and more AgTech AI companies should price it seriously. Third, media and raw material cost is one of the quietly decisive lines in any cost per kilogram model for a fermented ingredient, and that input chain is now being standardised by people who trade commodities for a living. Lower cost and less batch variability are good for the sector, but a founder whose in-house media recipe was on the differentiation slide should take it off. Finally, look at the sequence and the clock: pre-seed in March 2025, a customer case study, a co-developed product, a joint whitepaper, then a named partnership with a group of this size, all inside about eighteen months. Each step handed the corporate something it could show internally, which is how a small company gets a large one to sign.
