Anew Labs, the AI drug discovery company spun out of ByteDance, has closed a $290 million first external funding round at a post-money valuation of roughly $1.5 billion.

According to Yicai, Reuters and other outlets citing people familiar with the matter, the round was co-led by HSG (formerly Sequoia China), IDG Capital and Hillhouse Investment, with 5Y Capital as co-lead. Gaorong Ventures, Primavera Capital, Boyu Capital, Sino Biopharmaceutical and the Shanghai Future Industry Fund also participated.

ByteDance still holds about 56% of Anew Labs after the round, retaining control. As of the reports, neither ByteDance nor the lead investors had publicly commented on the details of the financing.

From In-House ByteDance Team to Standalone Company

Anew Labs began as ByteDance's internal AI for Science drug discovery team. Assembled around 2021, it brought together AI algorithm specialists and drug researchers, with roughly 50 core members.

In June 2026, ByteDance carved the team, its algorithm platform and its pipeline projects out into a separate entity. AI drug discovery differs sharply from internet content, advertising and recommendation businesses in R&D timelines, capital needs and regulatory requirements. Operating independently makes it easier to bring in outside capital and to build research and management systems suited to the biopharma industry.

Public filings show Anew Labs is headquartered in Shanghai, with work spanning biomolecular structure prediction, molecular dynamics simulation, antibody design optimization and drug discovery. The company has built out models and platforms including AnewFold, AnewSampling, AnewOmni, AnewDesign and AnewMind, and is advancing drug programs against targets such as IL-17 and IL-4R.

Launching AnewDDE, an Agent-Based Drug Discovery Platform

Alongside the funding news, Anew Labs unveiled AnewDDE, an agent-based drug discovery engine. The platform folds structure prediction, molecular dynamics, antibody design and scientific reasoning models into a single workflow.

Traditional drug discovery runs through target identification, structural analysis, molecular design, property prediction, synthesis and experimental validation, with each task typically handled by separate software and separate research teams. AnewDDE aims to use agents to coordinate those models and tools, letting the system break down tasks against a research goal, call the relevant compute modules and organize intermediate results.

The approach could cut the cost of switching between tools and wrangling data, but the platform's real value still hinges on whether its predictions hold up in the lab. Candidate molecules proposed by a model must still clear synthesis, activity, toxicity, pharmacokinetics and clinical trials.

So far Anew Labs has not released a full technical report on AnewDDE, nor disclosed external customer numbers or commercial revenue, and it has not detailed how far its lead programs are from clinical trials. Its platform capabilities and $1.5 billion valuation therefore remain to be validated by experimental results, partnerships and pipeline progress.

AI Prediction and Wet Lab Work Start Converging

As Anew Labs landed its large round, the experimental validation layer of the AI pharma supply chain saw a new partnership of its own.

On September 16, GenScript announced it was joining TuneLab, Eli Lilly's AI drug discovery platform, supplying wet lab services including protein expression, purification and characterization to participating companies. The services will help firms turn AI-selected protein or antibody sequences into physical samples and test their biological properties experimentally.

Lilly launched TuneLab in 2025. According to the company, the platform gives biotech firms access to selected AI models trained on Lilly's own R&D data, with the first batch covering antibody developability and small-molecule ADMET property prediction.

TuneLab relies on third-party hosting and federated learning. Participating companies can use the models without handing over raw data directly, and can feed experimental results back to improve the models over time. Lilly says the data behind the first models came from more than $1 billion in R&D spending.

GenScript's role covers part of the gap between AI prediction and experimental data generation. That points to a shift in how AI drug discovery platforms compete: less about supplying algorithms alone, more about linking models, data and standardized experimental capacity.

DNA Synthesis Firms Expand Capacity in Step

Twist Bioscience, an upstream supplier to AI drug discovery, has also raised its revenue outlook for fiscal 2026.

In its quarterly results, the company said it now expects fiscal 2026 revenue of $456 million to $457 million, up from an earlier forecast of $442 million to $447 million, with fourth-quarter revenue of $123 million to $124 million. Third-quarter fiscal 2026 revenue came in at $118.4 million, up more than 23% year over year.

In August, Twist also increased a planned stock offering from $250 million to $300 million, with proceeds earmarked for operations and capacity expansion.

AI models can generate large batches of candidate DNA, protein or antibody sequences at once, but those sequences still need synthesis and experiments to produce real data. As the volume of candidate designs grows, so may demand for DNA synthesis, protein expression and experimental testing.

That said, Twist's formal financial filings do not fully document the claim that AI orders have doubled three years running. That figure should not be read as a realized business result.

AI Pharma Competition Enters a Systems Phase

The Anew Labs round, GenScript's TuneLab tie-up and DNA synthesis capacity expansion map onto three distinct layers of AI drug discovery: computational design, experimental validation and basic materials.

Taken together, they suggest AI pharma is moving from a race between individual models toward coordination across the full R&D pipeline: models widen the search space for candidate molecules, experimental platforms generate reliable data, and pharma companies handle clinical development and commercialization.

But the road from an AI-generated candidate to an approved drug remains long. Even if models sharpen early-stage screening, they cannot remove the risks in toxicology studies, clinical trials and regulatory review. The metrics that matter for valuing AI pharma companies are still whether candidates reach the clinic, whether development timelines shorten, and whether success rates actually improve.

For Anew Labs, the things to watch next include formal confirmation of this round, the rollout of AnewDDE and customer traction, experimental and clinical results from its pipeline, and whether it can build a durable business model as a standalone company.