Getting started with Benchling AI

Benchling has a suite of AI-based productivity features. These features use large language models to intelligently interpret and generate natural language and make decisions.

Access to all AI-based features are managed by tenant-admins in the tenant admin console where each capability can be enabled or disabled individually. Your data is never used for cross-customer training or training by our AI model providers. For more information, see Data protection and security for AI at Benchling.

Benchling offers various AI agents, features and structure prediction models. AI agents are advanced AI powered capabilities that help you retrieve and capture data on Benchling. AI features provide lighter weight AI functionality to improve productivity. Structure prediction models leverage LLMs to help you visualize protein structures on Benchling. 

Capabilities marked preview or beta are available, and may be enabled via the tenant admin console using the steps below. Please keep in mind that these features are still a work in progress, and any feedback is welcome via the in-app feedback mechanisms.

Control Benchling AI access

Tenant Admins control Benchling AI capabilities through the Tenant Admin Console, where each capability can be enabled or disabled individually.

To control access to Benchling AI capabilities:

1. Click on your avatar in the Navigation bar to navigate to the tenant admin console 
2. Click the Settings tab
3. Click AI Settings in the menu 
4. Identify the feature you would like to edit
   • To disable for all users, click Disabled
   • To enable for all users, click Enabled for all users
   • To enable for specific users, click Enabled for select users, then use the search box to add all users who will have the feature enabled 
5. A Settings changed banner will appear, click Save to apply the changes 

These settings can be changed at any time by visiting the page again.

Note: Structure prediction models are not currently managed through the Tenant Admin Console. Contact Benchling Support to enable structure prediction models.

Onboard your team

Once your desired Benchling AI capabilities are enabled for the appropriate set of users, share key resources and best practices noted below with your team for them to get started. 

Best practices:

• Review our agent specific help center articles to find example use cases or prompts that your team can try
• Benchling AI agents use the permissions of the user. Before provisioning access, review your team's permission settings to ensure users have the right level of access
• Start with high-value, low-risk queries to build confidence
  • Example: Instead of a broad question like "Summarize all our cancer research," begin with a focused query such as "Summarize the results from the latest C9ORF72 antibody validation study."
• Incorporate into recurring workflows (weekly summaries, compliance checks)
  • Example: Instead of manually compiling a project update, use a prompt like "Generate a summary of our progress on the lead optimization phase of program X."
• Keep refining prompts for better accuracy
  • Example: A broad prompt like "Summarize our in vivo studies" can be refined to a more specific query: "Summarize the design of in vivo studies ST042 and ST043, including a table of key differences."
• Save useful queries for reuse
  • Example: If a query like "Write a report for how we engineered viral vector X, including the objective, methods, and results" proves effective, save the exact prompt so it can be easily adapted for future reports on other viral vectors.
• Try multiple queries in parallel

Key resources:

• Read about how other teams and organizations are applying Benchling AI
  • 3 labs share how Benchling AI is saving scientists time and toil
• Check out past Benchling AI trainings and discuss with others in our Community here

Customize Benchling AI with tenant-wide instructions

Deep Research, Compose, and Notebook Check support custom guidelines which provide additional admin defined context and rules that shape how the relevant AI agents and features work for your tenant.

Custom guidelines are managed by tenant admins in the Tenant Admin Console. To view or modify custom guidelines, open the Tenant Admin console’s settings page following the instructions above and navigate to the AI settings tab on the left hand side. 

Custom guidelines can be written in plain language and may contain rules to enforce, areas to focus on, explanations about which rules apply in which situations, etc. 

Customize Benchling AI with personalized instructions

In addition to admin-defined custom guidelines, individual users can define their own Custom Instructions for Benchling AI chat agents. Custom Instructions can be used to describe your role, share formatting preferences, or provide instructions to shape outputs, some examples are listed below: 

• I am an upstream process development scientist 
• I prefer bulleted summaries over long text paragraphs
• Always reference the Benchling Study name 

Custom Instructions can be updated at any time. Unlike custom guidelines, Custom Instructions apply only to your own conversations in Benchling AI and will affect other users in your tenant. They compliment custom guidelines and provide additional instruction to the LLM.

To add Custom Instructions: 

1. Use the sparkle icon in the navigation bar to open Benchling AI 
2. Click on the gear icon to open the Custom Instructions panel 
3. Use the textbox to enter any instructions that you'd like to see 

After you finish, your instructions will automatically save and you can return to the chat interface. 

Note: you can only create and save one set of instructions, you cannot save multiple sets or versions of instructions and switch between them to add different context for specific types of queries.    

Skills

Skills are reusable files that provide context and instructions for Benchling AI agents. Benchling’s agents can use skills to leverage learnings from one conversation and apply it reproducibly to future tasks.

Benchling Skills conform to the open Agent Skills standard.

There are two types of skills: 

• Benchling-authored skills that are written by Benchling and made available for use in your tenant 
• Custom skills written by you and uploaded in your AI dashboard

Custom skills 

If you have a prompt that you would like to re-use, you can use a third-party tool (like Claude or ChatGPT) to create a .zip or .skill file for upload into Benchling. Once you have the file, follow the steps below to upload into Benchling: 

1. Click on the AI icon in the navigation bar  
2. Click on the Settings icon in the AI dashboard, then click into the Skills tab of the AI Chat Settings menu
3. Click the Add button, this will open a window that allows you to upload your file 
4. Click Upload Skill 

If you need to update a custom skill, you would need to archive the existing skill and re-upload a new file.

Note: Custom skills can only be archived by their creator. Tenant admins do not have the ability to archive custom skills created by individual members of their teams. 

Benchling-authored skills 

Benchling-authored skills are auto installed by Benchling. Benchling-authored skills cannot be edited or deleted by users, Benchling will control all content and updates for these skills. 

The list of available Benchling-authored skills are below: 

• Report writing

See your Benchling AI usage

Open the Tenant Admin console’s settings page following the instructions above and navigate to the AI Usage tab on the left hand side. 

View your tenant's available AI credits and usage data and adjust the date period as needed. If you have questions about Benchling credits and usage, please reach out to Benchling Support.

Benchling Credits 

Benchling AI brings AI-powered features, agents, and models directly into the Benchling platform so scientists can accelerate their everyday R&D workflows. Some Benchling AI features are included with your subscription at no additional cost. Agents and models use a credit system — and to ensure every team can start using AI right away, credits are included with every Benchling subscription.

Features that don’t consume credits:

• SQL Writer 
• Notebook Checker

Agents and models that do consume credits: 

• Ask
• Deep Research
• Compose
• Data Entry Agent
• Structure prediction

The number of credits used per run is fixed by each agent and model, and depends on the complexity of the tasks being performed. See the current Rate Card and learn more here. 

Rate limits

Benchling AI has rate limit controls that help ensure compute capacity for all users at all times (and to prevent any unintended usage or abuse). These rate limits are set to default values that accommodate the vast majority of routine workflows. In some cases, such as a large volume of queries in a short time, you could run up against the rate limits. If you hit a rate limit, you'll see a notification that reads: "You have exceeded the daily usage limit for AI features. Please try again later."

If your team is hitting rate limits frequently, or if hitting a rate limit is blocking work, an admin can contact Benchling Support to request an increase. Adjustments can be made for specific users or for the entire tenant depending on the situation.

Note: rate limits are independent from Benchling Credits, which means you or other users on your tenant could hit a rate limit even when credits are still available. Benchling Support can adjust rate limits, but they cannot adjust your Benchling Credits balance. Adding monthly credit capacity requires a purchase order.