Answers from the documentation; checking which model is on duty
Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.
Grounded in the docs - follow a source before acting on it.
Pierre Fabre Uses XtalPi’s AI and Robotics for Drug Discovery - OpenSmartRoute
Pierre Fabre Uses XtalPi’s AI and Robotics for Drug Discovery
Pierre Fabre and XtalPi announced a partnership to discover new small-molecule drugs. XtalPi provides AI and robotics, while Pierre Fabre handles selection and development.
Key points
Partnership announced on November 9, 2026, for drug discovery
XtalPi will handle de novo design of new molecules
Focus on challenging targets in oncology and rare diseases
Agreement includes upfront and milestone payments
Why it matters: Accelerates discovery of high-quality drug candidates in areas with unmet medical needs.
By OpenSmartRoute editorial · written through the router by llm-small
From Unite.AI - “Pierre Fabre Taps XtalPi’s AI and Robotics for Drug Discovery”
Introduction - Partnership announcement and its purpose
Laboratoires Pierre Fabre and XtalPi announced a new partnership on November 9, 2026. The goal is to discover new small-molecule drugs for various medical conditions. This collaboration combines Pierre Fabre’s expertise in medicine with XtalPi’s advanced AI and robotics technology. The partnership aims to find promising drug candidates faster and with higher quality. It focuses on challenging areas where drug discovery has been difficult in the past.
This partnership marks a significant step in using artificial intelligence (AI) to improve how new medicines are developed. It shows how traditional pharmaceutical companies are now working closely with AI technology providers. The purpose is to accelerate the process of finding effective drugs for unmet medical needs. By combining their strengths, Pierre Fabre and XtalPi hope to make drug discovery more efficient and successful.
Details of the collaboration - Roles of Pierre Fabre and XtalPi
In this partnership, XtalPi provides its integrated AI capabilities and robotics platform. Its role is to create new molecular structures from scratch. This process is called de novo design, meaning designing molecules that have never been made before. XtalPi will generate novel small-molecule therapeutics aimed at targets that are difficult to drug with traditional methods.
Pierre Fabre’s role is to select the most promising molecules created by XtalPi. It will use its expertise in disease biology, discovery, translational research, and clinical development. Pierre Fabre will decide which molecules to advance into further testing and development. It will also handle the later stages of clinical trials and commercial production. The partnership ensures that Pierre Fabre maintains control over the decision-making process after the molecules are designed.
Focus areas - Oncology, dermatology, rare diseases, and hard-to-drug targets
The collaboration covers multiple research programs in different medical areas. The main focus is on oncology, which involves cancer treatments. There is also potential work in dermatology, which deals with skin conditions. Rare diseases, which affect small patient groups, are another key area. These conditions often lack effective treatments, making them important targets for new drug discovery.
Andrej Karpathy used Claude's Opus 5.5 model to create HackerTube, a curated video site from 20 years of Hacker News comments in a weekend.
The partners are especially interested in targets that have been difficult to drug in the past. These are often considered hard-to-drug because they resist traditional approaches. Focusing on these challenging targets aims to address significant unmet medical needs. The goal is to develop drugs that can treat conditions where current options are limited or ineffective.
Technologies involved - AI-driven molecular design and robotics platform
XtalPi’s platform combines artificial intelligence with robotics. Its AI algorithms are used to design new molecules based on known scientific data. The platform can generate many candidate molecules quickly. Robotics automates parts of the process, such as synthesizing and testing molecules in the lab.
This integrated approach allows for rapid iteration. The AI suggests promising molecules, and the robotics platform tests them. The results then feed back into the AI models, improving future designs. This cycle speeds up the discovery process and helps identify high-quality candidates more efficiently than traditional methods.
Process flow - From molecule design to candidate selection
The process begins with XtalPi’s AI creating new molecular structures. These molecules are designed to target specific disease mechanisms. The AI considers factors like drug-likeness, stability, and potential effectiveness. Once designed, the molecules are produced and tested using robotics.
The testing phase involves evaluating the molecules’ properties and interactions. Data from these tests are used to refine the AI models. The most promising molecules are then handed over to Pierre Fabre. Its team reviews the candidates based on biological relevance and potential for development. Pierre Fabre selects the best molecules to move forward into preclinical and clinical testing.
Scope and scope of research - Multiple programs and therapeutic areas
The partnership covers several research programs across different therapeutic areas. The focus is primarily on oncology, where the need for new treatments is urgent. Other areas, such as dermatology and rare diseases, are also included. The scope allows XtalPi’s platform to be tested in various settings, providing valuable feedback.
Multiple programs mean that the partnership can evaluate the platform’s performance across different targets and disease types. This broad scope increases the chances of discovering effective drugs. It also helps both companies learn how best to apply AI and robotics in different research contexts.
Financial terms - Payments and economic participation
The agreement includes an upfront payment from Pierre Fabre to XtalPi. There are also multiple milestone payments, which are made when specific research goals are achieved. Additionally, the partnership involves downstream economic participation, meaning XtalPi may share in the financial success of any drugs that reach the market.
These financial arrangements reflect the shared risks and rewards of the collaboration. The upfront payment helps fund initial research activities. Milestone payments motivate the partner companies to meet development goals. The economic participation aligns both parties’ interests in the success of the drug candidates.
Background - AI in drug discovery and XtalPi’s platform
Artificial intelligence has become an important tool in drug discovery. It helps identify promising targets, design new molecules, and predict how drugs will behave. AI can analyze vast amounts of scientific data faster than traditional methods.
XtalPi’s platform combines AI algorithms with robotics to automate the early stages of drug discovery. It can generate and test many molecules in a short time. This approach aims to reduce the time and cost of bringing new drugs to market. It also increases the chances of finding effective treatments for difficult targets.
Why it matters - Impact on drug development speed and quality
This partnership could significantly speed up the process of discovering new drugs. Using AI and robotics allows for faster design and testing of molecules. It also improves the quality of candidates by focusing on those most likely to succeed in clinical trials.
Focusing on hard-to-drug targets addresses unmet medical needs. Developing drugs for these conditions can lead to breakthroughs in treatment. Overall, this approach aims to make drug discovery more efficient, less costly, and more successful in delivering new medicines.
How OpenSmartRoute helps
Routing requests through OpenSmartRoute allows teams to efficiently manage AI and model requests. It sends each request to the best-fit model or agent from a catalog defined by the team. The system scores each candidate based on quality, cost, speed, and safety, with team-set weights.
For a project like this, OpenSmartRoute can route molecule design requests to the most suitable AI models or tools. It can enforce rules, such as keeping sensitive data on-premises or respecting cost limits. The system learns from outcomes, directing more traffic to successful models and less to those that fail. This helps teams optimize their AI workflows for drug discovery projects.
How it compares - what existed before, what this changes and what stays the same
Before this partnership, drug discovery relied heavily on traditional methods. Researchers used laboratory experiments and manual screening to find promising molecules. These processes took years and cost a lot of money.
AI has been used in some parts of drug discovery. It helped identify targets and predict molecule behavior. But AI tools were often separate from robotics platforms. They did not fully automate the design and testing process.
XtalPi’s platform combines AI algorithms with robotics. It automates the creation and testing of molecules. This integration allows for faster iteration and higher throughput. It can generate many molecules and evaluate them quickly.
This partnership shifts the focus to using AI and robotics together for de novo design. It emphasizes creating new molecules from scratch, not just screening existing ones. The goal is to find high-quality candidates faster and more efficiently.
What stays the same is the overall goal of discovering effective drugs. The importance of understanding disease biology and translating findings into clinical candidates remains. The partnership enhances existing capabilities, not replaces them.
Traditional drug discovery methods still play a role in later stages. Human expertise in biology and clinical testing remains essential. The new approach mainly accelerates early discovery and design phases.
Questions this leaves open - what the source does not say and how a reader can check it
The source does not specify the size of XtalPi’s AI models or robotics platform. It also does not detail the number of molecules they can generate or test per day. Knowing these numbers would help assess the platform’s scale.
It is unclear how the AI models are trained or validated. The source does not mention the data sources or the accuracy of the predictions. Understanding the validation process is important for evaluating reliability.
The source does not specify the exact targets or disease areas beyond oncology and potential dermatology. It is unknown whether the platform can handle other therapeutic areas or rare diseases. Checking the company’s published research or technical documentation could clarify this.
Details about the licensing or access to XtalPi’s platform are not provided. It is unclear whether other partners or internal teams can use the platform for different projects. Contacting the companies or reviewing their product offerings might provide insights.
The source does not mention the timeline for project milestones or the expected duration of the partnership. Knowing the schedule would help understand how quickly results might emerge.
It is also not clear how success will be measured. Will the number of molecules progressing to clinical trials be a key metric? Or the reduction in discovery time? Clarifying these criteria would help evaluate progress.
Finally, the source does not discuss potential challenges or limitations. For example, AI-generated molecules may still face hurdles in synthesis or toxicity. Investigating these issues through scientific literature or industry reports can provide a balanced view.
