--- title: "ByteDance Spins Out Drug Lab With $290M Round at $1.5B Valuation" date: 2026-10-05 category: Capital & Markets site: NeuroAI canonical: https://neuroai.site/a/na-capital-anew-labs-bytedance-spinout language: en --- # ByteDance Spins Out Drug Lab With $290M Round at $1.5B Valuation > Anew Labs, carved out of ByteDance, raised US$290 million at a US$1.5 billion valuation to build AI-native drug discovery. A pharmaceutical bet that began as an internal experiment inside a short-video company is now standing on its own balance sheet. In mid-September, a Shanghai outfit few outside the industry had heard of closed one of the year's largest early-stage AI-biotech rounds — and it did so still majority-owned by the parent that created it. Anew Labs (新生实验室), spun out of ByteDance earlier in 2026, has completed a US$290 million first external financing at a reported US$1.5 billion post-money valuation. Reuters, citing people familiar with the transaction, reported the deal on September 16; the company and its investors did not comment. ByteDance retains a 56 percent stake, keeping the lab inside its orbit while giving it a separate capitalization table and external governance. ## From prediction to design The split is the interesting part. ByteDance had quietly built an AI-for-Science (AI4S) team from 2021, led by Liu Kai (刘凯), blending algorithm researchers with veteran drug developers. What started with structure prediction has moved into molecular design — the harder, more valuable step of proposing new candidates rather than merely describing existing proteins. Anew Labs describes an AI-native discovery platform built on transformer networks that design novel synthetic proteins, simulate how molecules dock, and forecast drug-target behaviour. Combined with a high-performance computing layer and automated biochemical screening, the pitch is that a client can move from disease profiling to a validated lead candidate in weeks rather than years. The product roster reads like a pipeline of pipelines: - **AnewFold** — predicts protein and molecular-complex structures. - **AnewOmni** — generates and designs small molecules, peptides and antibody binders. - **AnewDesign** — handles antibody design and optimization. - **AnewDDE** — links structure prediction, molecular design, binding-affinity ranking and scientific reasoning, then feeds experimental results back into the next design loop. - **AnewMind** — supports scientific reasoning and go/no-go decisions across the discovery chain. The disclosed early pipeline targets include the IL-17 family (IL17AA/AF/FF) and IL-4R, plus two undisclosed programmes. ## Who wrote the cheque The round was co-led by HSG (the firm formerly known as Sequoia Capital China), IDG Capital and GL Ventures, with 5Y Capital as co-lead and participation from Gaorong Ventures, Primavera Venture Partners, Boyu Capital, SBP Group and the state-backed Shanghai Future Industries Fund (上海未来产业基金). That mix — top-tier venture, sovereign-linked local capital and strategic pharma money — is the kind of cap table that signals investors are buying a platform, not a single molecule. In the conversions the rules of this market imply: US$290 million is about ¥2.06 billion (≈ HK$2.27 billion), and the US$1.5 billion valuation is roughly ¥10.65 billion (≈ HK$11.7 billion). Those are approximation markers, not audited figures. ## A wider wave, not an isolated bet Anew Labs is the headline, but it sits inside a broader surge. China's Ministry of Industry and Information Technology and ten other ministries published the 15th Five-Year (十五五) implementation plan for the pharmaceutical industry in September 2026, with a dedicated column on cultivating high-value AI-drug (AI制药) application scenarios — targeting discovery, molecular design and optimization. The money is following the policy. Industry data cited by Chinese financial media shows AI-drug deals in 2026 reached 57 transactions across 44 companies with more than ¥15.76 billion disclosed — already more than the combined total of 2023, 2024 and 2025. Globally, the comparison is Alphabet's Isomorphic Labs (born from DeepMind), which closed a US$2.1 billion Series B in May 2026 at a valuation reported near US$15–20 billion despite having no mature clinical pipeline of its own. The pattern is the same on both sides of the world: investors are pricing the model and the wet-lab loop, not the drug. ## What readers can do now - If you follow biotech investing, separate "AI-native discovery platform" valuations from traditional pipeline risk — Anew Labs' US$1.5 billion tag is a technology premium, not a read on any single drug's odds. - Watch for the company's first clinical-candidate nomination; the honest test of this bet is whether its models produce molecules that survive toxicology and trials, not benchmark scores. - Track the Shanghai Future Industries Fund and similar state-linked vehicles — their presence in private rounds is a useful signal of where Beijing wants AI capability commercialised. ## Honest limitations The financing was disclosed through Reuters reporting based on sources; neither ByteDance nor Anew Labs publicly confirmed the terms, and the parties declined to comment, so the US$290 million amount, the US$1.5 billion valuation and the 56 percent stake are journalistic figures rather than filed ones. The fastest disclosed pipeline programme remains at the pre-clinical stage, so the valuation rests on platform potential and ByteDance's compute support, not on clinical validation. No independent audit of the discovery platform's hit rates has been published. This is a capital-markets and strategy account, not investment advice, and the long timeline of drug development means outcomes are years from settlement. --- Published by NeuroAI (https://neuroai.site/) — https://neuroai.site/a/na-capital-anew-labs-bytedance-spinout Free to quote with attribution and a link to the original.