Ask DeepSeek AI

Write one prompt and run DeepSeek on every row of your CSV or Excel file. Free in Datablist with your own DeepSeek API key.

Start for free

No credit card needed

Free With Your Own KeyCSV and Excel ImportBulk AI Without Writing Code

Overview

At a Glance

Ask DeepSeek AI runs one prompt on every row of your file. Use values from each row inside the prompt, and Datablist handles rate limits, retries and a status per row.

What you give

Your DeepSeek API Key: Pasted into the settings, used only for your runs
A Prompt: One prompt for the whole file, with columns inserted by typing a slash
Optional System Prompt: Sets the role or the answer format

What you get

A Result Column: The text returned by DeepSeek for each row
Structured Outputs: Optional: several named columns from one prompt
A Status per Row: See which rows succeeded and rerun the failed ones

How it's billed

Free in Datablist: No Datablist credits are used
Billed by DeepSeek: Usage is charged to your own DeepSeek account
Growth for API and MCP: Running through the API or MCP needs the Growth plan

Sample Data

What Your Results Look Like

Input: Company Name. Output: Result.

Sample input and output.
Your data	Added by Ask DeepSeek AI
Company Name	Result
Stonebridge Capital	Financial services
Copperridge Energy	Energy
Lumenpath Health	Healthcare
Boulangerie Lefèvre	Bakery
Rheintal Elektro	Electrical

Example

Example: Classify companies by industry

A prompt that works on every row of a company list. Open it to see the exact wording and how the column is inserted.

Show the example prompt

Prompt

Classify this company by industry. Return only the industry name:

# Company Name
{{Company Name}}

System prompt (optional)

Return just the result. No instruction, no comment.

Type a slash in the prompt field and choose the Company Name property to insert it. Datablist replaces it with the value from each row.

Plans

Free to run, with the right plan

Ask DeepSeek AI does not use Datablist credits, so there is no cost to estimate. Here is what you need.

Datablist credits: None. This enrichment is free to run.
DeepSeek usage: DeepSeek bills your own account for API usage. Fees come out of your topped-up balance or your granted balance, with the granted balance used first.
Running in the app: No paid Datablist plan needed. Add your API key, write your prompt and run it.
Running through the API or MCP: Needs the Growth plan. The enrichment stays free of credits, but API and MCP access come with Growth.

Try for free

No credit card needed View pricing

Good to know

Your DeepSeek account needs a balance for the run to work. Top it up in the DeepSeek Platform before you start a large file.

The Process

How it Works

Write one prompt, point it at your data and run it on every row.

Step 1

Add your key and prompt

Paste your DeepSeek API key, then write one prompt for the whole file. Type a slash to insert a column such as the company name. Add a system prompt to set the format, for example to return just the result.

Step 2

Test on a small batch

Run the prompt on the first 10 rows and read the results. Adjust the wording, the system prompt or the model until the answers match what you expect, before you spend time on the full file.

Step 3

Run the full list

Datablist sends one request per row, manages throttling and retries, and tracks a status for each row. Filter the failed rows and rerun only those.

Start for free

Trusted by fast-growing startups and Fortune 500 teams

Workflows

Templates

Jumpstart your lead generation and outreach with pre-configured enrichment recipes.

Classify as B2B or B2C business

Use a company description to classify it as B2B or B2C.

Analyse if a company is a SaaS company

Use the company description and return whether it is a SaaS business.

Checks if company is an Ideal Customer Company

Analyze a company description and website text and identify if it fits your Ideal Customer Profile.

Guides

Frequently Asked Questions

Answers to common questions about keys, billing, limits and results.

: Yes. Datablist runs your prompt through your own DeepSeek account, so you paste your API key into the settings. Your account must be able to call the API, which can need credits or billing.
: Datablist does not charge credits for it. DeepSeek bills your own account for API usage. Fees come out of your topped-up balance or your granted balance, with the granted balance used first.
: Not to run it in the app. If you want to run it through the API or MCP, you need the Growth plan.
: Yes. Datablist is built for large collections, so you can import a big file and run the prompt on every row, within your plan’s collection limits. Test a small sample before running the full list.
: Datablist manages throttling and retries failed calls. If your account still hits limits, lower Max Concurrent Calls and switch on Run in Async.
: Yes. Datablist stores a status for each row. Filter the rows with errors, fix what is needed and rerun only those rows.
: Yes. Switch on Define outputs format and name each output, for example a category, a summary and a score. Datablist stores each one in its own column.
: It can be, because you pay DeepSeek directly per token and can choose a model by price. Test on a small batch to check that the answers are good enough for your task.
: Sign in to the DeepSeek Platform and top up your balance. Then open API keys, create a new key and copy it somewhere safe. Paste it into the DeepSeek API Key field in the enrichment.
: Type a slash in the prompt field and choose the property from the list. You need to type the slash and select the property, so the value is inserted for each row.
: It gives the model standing instructions, such as its role or the answer format. A line like Return just the result. No instruction, no comment. keeps answers short.
: Pick a model in the DeepSeek Model setting. To use a model that is not in the list, select Other, then type its model code in the Model code setting, exactly as written on DeepSeek's models and pricing page.
: It controls how many requests run in parallel. The setting defaults to 10, or 20 in async mode. Lower it if you see rate limit errors.
: Yes. Use the chevron next to the run button to process the first 10 or 100 items and check the results before you run the full list.
: The text returned by the model for each row. Rows that failed show an error in the run status, so you can filter and rerun them.
: Send only the columns needed for the task, such as a product description, lead note, company summary, or support message. Smaller prompts are easier to review and cheaper to process.

Integrations

Related Enrichments

Pair Ask DeepSeek AI with complementary data tools in Datablist.

Ask ChatGPT/OpenAI

Run a ChatGPT prompt on every row.

Ask Claude AI

Run Claude on every row.

AI Agent

Research each row on the web with an AI agent.

See All Enrichments

Free Tools

Try It Free First

Free versions of related tools, to try on a few rows before you scale up.

Free Company Name Cleaner

Normalize company names: legal suffixes stripped, casing fixed.

Free Email Finder

Find a work email from a name and a company. Free to try.

Free Domain Finder

Paste company names and find their websites. Free, no credit card required.

See All Free Tools

How to use Datablist's Ask DeepSeek AI

Run Datablist's Ask DeepSeek AI on your own list in 8 steps, no code needed. The settings reference and answers to common questions follow the guide.

Easy to UseFully No-Code38,000+ Users

Step-by-Step Guide

Part 1: Get your DeepSeek API key

Follow DeepSeek's own instructions: DeepSeek API docs.

Top up your balance. Usage is deducted from your topped-up balance or your granted balance, and the granted balance is used first.

Step 2: Create and Copy Your Key

Open API keys and create a new key.

Copy the key and store it somewhere safe.

Part 2: Run the enrichment in Datablist

First, sign up for Datablist.com

Then, upload your CSV or Excel file with the rows you want to run the prompt on. Create a new collection and import your file with the CSV editor.

Step 4: Navigate to the Enrichment

Click Enrich in the top menu and search for Ask DeepSeek AI.

Ask DeepSeek AI - Navigate to the Enrichment

Step 5: Paste Your API Key

In the drawer that opens, paste your DeepSeek API key in the DeepSeek API Key field.

Datablist sends the key with your run to call DeepSeek. Do not share it with others.

Step 6: Write Your Prompt

Write one prompt for all your rows. Type a slash and choose a property to insert its value, for example the company name.

Show an example prompt

Classify this company by industry. Return only the industry name:

# Company Name
{{Company Name}}

The prompt must include at least one row variable. Type {{ or / and select a property; all your text columns can be used.

Optionally add a system prompt to guide the answer, for example: Return just the result. No instruction, no comment.

Step 7: Choose the Output Columns

DeepSeek returns the answer in a Result column. To get several fields from one prompt, switch on Define outputs format and name each output. Then choose where to store each result: create new properties or map existing ones.

Click Continue to outputs configuration, then click the ⊕ icons to add the output properties to your collection.

Each structured output has a name, description and type: Text, Long Text, URL, Number, DateTime or Checkbox. Define an output for each field your prompt requests. This option is available on compatible models.

Ask DeepSeek AI - Choose the Output Columns

Step 8: Test on a Few Rows, Then Run

Run a sample first. Click the chevron on the right side of the run button to choose how many items to process:

Run on first 10 items: Good for checking results before committing
Run on first 100 items: Useful for validating larger samples
Run on all view items: Process your entire list (or view if you've enabled filters)

Check the results, adjust your prompt, then run the full list. Switch on Run in Async in the Run Settings to let Datablist handle automatic background retries.

Review the run-status column, filter rows with errors, fix the input or settings, and retry only those rows. You can filter and export the resulting collection.

Ask DeepSeek AI - Test on a Few Rows, Then Run

Use cases

Code Explanation or Generation

If your rows contain code snippets, ask DeepSeek to explain them. Or, generate simple code based on parameters in other columns.

Data Cleaning

Use DeepSeek to correct text cases, fix grammar and typos, or standardize messy text fields across thousands of rows.

Text Summarization

Generate concise summaries from long descriptions, articles, or feedback found in each row.

Language Translation

Translate phrases or sentences from one language to another using DeepSeek's language capabilities.

Content Categorization

Automatically assign categories or tags to items (like products or articles) based on their text descriptions.

Information Extraction

Pull out specific pieces of information (like names, dates, company mentions) from unstructured text in your rows.

Settings Reference

A lookup for every setting, input and output. The step-by-step guide above covers everything you need to get started.

Settings

Settings
DeepSeek API Key	Text	Your DeepSeek API Key. Available at platform.deepseek.com/api_keys. Enabled when: `noAPIKey==false`
Prompt	RichText	Prompt to be used to call DeepSeek. Type a slash and choose a property to add it as a variable. Ex: 'Translate the following text in Italian' followed by the Text property, or 'Return the first name for' followed by the Name property.
System Prompt (Optional)	LongText	The system prompt gives instructions for the chatbot to guide its responses or actions. Example 1: Your job is to list relevant keywords to a company's prospecting. You will help me by... Example 2 (to avoid useless text): Return just the result. No instruction, no comment.
DeepSeek Model	Text	Choose the model to use. See DeepSeek's models and pricing page to learn more.
Max Concurrent Calls	Number	Maximum number of parallel calls. The setting defaults to 20 in async mode, 10 otherwise. Reduce if you hit rate limit errors. Enable Run in Async in the Run Settings (last step) to let Datablist handle automatic background retries.
Model code	Text	Write the model code name, as written on DeepSeek's models and pricing page. Select Other in DeepSeek Model first. Enabled when: `llmModel==custom`
Define outputs format	Boolean	Specify the format for the outputs. This enables multiple outputs and better data types.
Outputs Format	MultipleValues	Define each expected output. Outputs must match your prompt intent. Enabled when: `defineOutputsFormat==true`
Outputs Format → Output Name	Text	Example: Company Name, First Name, Result Enabled when: `defineOutputsFormat==true`
Outputs Format → Output Description	Text	Context for the output. It helps the model to understand what to return. Ex: The cleaned name; The email content, should start with XX; Enabled when: `defineOutputsFormat==true`
Outputs Format → Output Type	Text	Default to Text. Enabled when: `defineOutputsFormat==true`

Inputs

No fixed fields are declared. Available fields may depend on your settings.

Outputs

Outputs
Result	LongText	The text returned by the LLM.

Run ChatGPT on Every Spreadsheet Row

How to run ChatGPT on Excel and CSV rows

Write Prompts to Analyze Your Data

How to Write a Prompt to Analyze and Classify Data (Full Guide)

Score Your Leads with AI

How to Score Leads With AI

FAQ

Frequently Asked Questions

Answers to common questions about keys, billing, limits and results.

: Yes. Datablist runs your prompt through your own DeepSeek account, so you paste your API key into the settings. Your account must be able to call the API, which can need credits or billing.
: Datablist does not charge credits for it. DeepSeek bills your own account for API usage. Fees come out of your topped-up balance or your granted balance, with the granted balance used first.
: Not to run it in the app. If you want to run it through the API or MCP, you need the Growth plan.
: Yes. Datablist is built for large collections, so you can import a big file and run the prompt on every row, within your plan’s collection limits. Test a small sample before running the full list.
: Datablist manages throttling and retries failed calls. If your account still hits limits, lower Max Concurrent Calls and switch on Run in Async.
: Yes. Datablist stores a status for each row. Filter the rows with errors, fix what is needed and rerun only those rows.
: Yes. Switch on Define outputs format and name each output, for example a category, a summary and a score. Datablist stores each one in its own column.
: It can be, because you pay DeepSeek directly per token and can choose a model by price. Test on a small batch to check that the answers are good enough for your task.
: Sign in to the DeepSeek Platform and top up your balance. Then open API keys, create a new key and copy it somewhere safe. Paste it into the DeepSeek API Key field in the enrichment.
: Type a slash in the prompt field and choose the property from the list. You need to type the slash and select the property, so the value is inserted for each row.
: It gives the model standing instructions, such as its role or the answer format. A line like Return just the result. No instruction, no comment. keeps answers short.
: Pick a model in the DeepSeek Model setting. To use a model that is not in the list, select Other, then type its model code in the Model code setting, exactly as written on DeepSeek's models and pricing page.
: It controls how many requests run in parallel. The setting defaults to 10, or 20 in async mode. Lower it if you see rate limit errors.
: Yes. Use the chevron next to the run button to process the first 10 or 100 items and check the results before you run the full list.
: The text returned by the model for each row. Rows that failed show an error in the run status, so you can filter and rerun them.
: Send only the columns needed for the task, such as a product description, lead note, company summary, or support message. Smaller prompts are easier to review and cheaper to process.

At a Glance

What you give

What you get

How it's billed

What Your Results Look Like

Example: Classify companies by industry

Prompt

System prompt (optional)

Free to run, with the right plan

Good to know

How it Works

Add your key and prompt

Test on a small batch

Run the full list

Templates

Related Tutorials

Frequently Asked Questions

Related Enrichments

Try It Free First

Step-by-Step Guide

Part 1: Get your DeepSeek API key

Step 1: Sign In and Top Up Your Balance

Step 2: Create and Copy Your Key

Part 2: Run the enrichment in Datablist

Step 3: Sign Up & Upload Your Data

Step 4: Navigate to the Enrichment

Step 5: Paste Your API Key

Step 6: Write Your Prompt

Step 7: Choose the Output Columns

Step 8: Test on a Few Rows, Then Run

Use cases

Code Explanation or Generation

Data Cleaning

Text Summarization

Language Translation

Content Categorization

Information Extraction

Settings Reference

Settings

Inputs

Outputs

Related Articles

Frequently Asked Questions