Cohere is in advanced talks to raise $2 billion to $3 billion at a $20 billion valuation, according to a September 11 report by The Globe and Mail. The Canadian government and existing backers are expected to participate, Germany is also in talks and non-government investors are expected to provide the majority. The financing could close as early as next week, but the terms are not final.
The valuation predates the September talks. On April 24, when Cohere announced its planned combination with Germany's Aleph Alpha, Axios reported that the combined company would be valued at about $20 billion after a concurrent Series E. Schwarz Group committed €500 million, or roughly $600 million, in structured financing and was identified as the lead investor in that round.
September is a capital-depth test at a price already telegraphed in April. The late-stage investor's question is whether private and sovereign backers will provide another $2 billion to $3 billion on terms consistent with that structure, and what operating scale now supports a price almost three times Cohere's prior $7 billion mark.
The public revenue denominator is old, but it is too important to omit
The latest public ARR figure cited in current reporting is roughly $240 million for Cohere in 2025, according to TechCrunch. Dividing $20 billion by $240 million gives about 83 times ARR. The ratio is deliberately mismatched: the numerator belongs to the planned combined Cohere-Aleph Alpha business after financing, while the denominator is Cohere-only ARR from the prior year. Aleph Alpha also brings assets and institutional relationships absent from Cohere's 2025 ARR.
The calculation is still useful as a scale check because it shows how far the headline valuation sits above the last public commercial base. An investor needs substantial revenue growth, meaningful Aleph Alpha economics, or strategic value beyond software revenue to make the price look less extreme. A proper current multiple needs combined ARR, growth, gross margin and cash burn; the public record still stops at the older Cohere-only revenue base.
Cohere's strategy helps explain why governments and strategic investors are interested. The company has positioned itself around secure enterprise and sovereign AI, and the Aleph Alpha combination deepens its German and European institutional footprint. Canada has already offered up to C$240 million of support under its AI Compute Challenge and has explored government use of Cohere technology. Those relationships can improve compute access, distribution and procurement credibility. Revenue, margins and signed customer spend still determine how much commercial value sits underneath that strategic positioning.
The investor mix now carries more information than another repetition of the $20 billion figure. A majority-private close with Canadian and German participation would show market capital accepting the sovereign-AI thesis at the April price. Heavy dependence on public-sector money would tell a different story about private risk appetite.
The treatment of Schwarz's roughly $600 million April commitment is also unclear: the September report leaves open whether it sits inside the $2 billion-$3 billion target or alongside it. Final pre- or post-money convention, security type and investor allocation remain private, leaving dilution unquantified.
A $2 billion-$3 billion close would change Cohere's capital depth far more than its price discovery: $20 billion was already in the market. The 83-times comparison is useful because it exposes the missing denominator, not because it supplies a current multiple. Cohere's current combined ARR, margins and burn will decide whether the April price looks merely ambitious or extraordinary; September's financing tells investors how much capital is willing to underwrite that bet.