AI & Models · Integration
DeepSeek
Add DeepSeek to your product for your customers, and give your AI agents governed access to it.
DeepSeek serves an OpenAI-compatible API at api.deepseek.com, so existing SDKs work by changing the base URL and key, and the optional /v1 path segment refers to API compatibility rather than a model version. Two models matter: deepseek-chat and deepseek-reasoner, and the reasoner returns a reasoning_content field alongside content which must not be fed back into the next messages array, otherwise the request is rejected. Usage on every response splits prompt tokens into prompt_cache_hit_tokens and prompt_cache_miss_tokens because context caching is automatic, so keeping a stable prefix at the front of a prompt directly lowers cost. The surface is narrower than it looks: there is no embeddings endpoint, so vector work needs another provider. GET /user/balance reports remaining credit, which is worth checking before a batch job since exhausted balance fails requests rather than queueing them. fastn holds each customer's API key, keeps the connection current and controls which agents may spend their credit.
In your product
Embedded for your customers. Per-tenant auth, no per-customer code, maintained by fastn.
Let a customer connect their own DeepSeek account so inference is billed to their key and their data agreement.
Offer DeepSeek as a model choice alongside other providers using the same OpenAI-compatible call path.
Use deepseek-reasoner for tasks that need visible working and store the reasoning separately from the answer.
Show remaining balance and cache hit rates so a customer can see what their prompts actually cost.
For your AI agents
Governed, audited access for the agents you build, through the MCP server.
Let an agent run an approved prompt template against deepseek-chat and return the output, with token spend and cache hits audited per tenant.
Have an agent check the account balance and stop a batch job that would exhaust it.
Allow an agent to call chat completions but not to read or rotate the underlying API key.
Example prompt
Summarise this contract with deepseek-reasoner and show me how much of the prompt was cached.
Set up DeepSeek in 4 steps
- 01Enable the DeepSeek connector from your fastn dashboard.
- 02Have each customer authorise their own DeepSeek account, so calls run under their credentials rather than a shared key.
- 03Decide which models, requests and outputs your product needs, map those fields, then enable the actions and triggers you want.
- 04Call it from your product and expose it to your agents through the same governed connection.
Why teams use the DeepSeek integration
What you get by embedding it with fastn instead of building it yourself.
- Ship a DeepSeek integration without building it. Your customers connect their own DeepSeek account inside your product and work their models, requests and outputs there, with no per-customer code on your side.
- Handle the part that actually costs time: per-customer keys, quotas and cost attribution matter more than schema here, because every call is billed. fastn owns the auth, token refresh, rate limits, pagination and breaking-change fixes, so a DeepSeek update is not your on-call problem.
- One integration serves your product and your agents. The same governed DeepSeek connection powers in-product features and gives AI agents scoped, audited access, so you give your product and your agents a model call each customer pays for themselves without wiring it twice.
Used by these teams
Compare with
Anthropic Claude
OpenAI
Azure OpenAIOften used alongside
Tools the same teams tend to run next to DeepSeek, across other categories.
Snowflake
Airtable
PostgreSQL
AmplitudeDeepSeek integration FAQ
How do I add a DeepSeek integration to my product?
Enable the DeepSeek connector in your fastn dashboard, then let each customer authenticate their own DeepSeek account. fastn handles the OAuth flow, token storage and refresh per tenant, so there is no DeepSeek client code in your app and no per-customer branch in your codebase. Setup is 4 steps.
Do my customers each connect their own DeepSeek account?
Yes. Every connection is scoped to the individual customer, so each authorises their own DeepSeek account and only ever sees their own models, requests and outputs. That per-tenant isolation is the point of an embedded integration: you support the long tail of customer setups without maintaining an integration per customer.
Can AI agents use this DeepSeek integration?
Yes. The same connection is exposed to your agents through the fastn MCP gateway, with permissions scoped per tenant and every call audited. Let an agent run an approved prompt template against deepseek-chat and return the output, with token spend and cache hits audited per tenant.
Who maintains the DeepSeek integration?
fastn does. When DeepSeek changes an endpoint, deprecates a field or alters its auth, the fix lands in the connector rather than in your backlog, and your customers' connections keep working.
Whose DeepSeek API key and quota does each call use?
Each customer authorises their own DeepSeek account, so usage, rate limits and cost land on the customer that caused them. You are not metering a shared key and re-billing it, and one heavy customer cannot exhaust another's quota.
Are inputs and outputs auditable?
Yes. Every call is logged per tenant with the call, the input and the result, so an output can be traced back to what produced it. That matters more here than in most integrations, because a generated answer or an extracted field cannot be reconstructed from the request alone.
What can I build with the DeepSeek integration?
A common starting point: let a customer connect their own DeepSeek account so inference is billed to their key and their data agreement. Teams also use it for the other use cases listed above, and expose it to agents for governed reads and writes.
How much does the DeepSeek integration cost?
It is included. Pricing is based on connected accounts, not on how many connectors you enable, so adding DeepSeek does not change your per-connector cost. You can start free with 3 connected accounts.
Add DeepSeek to your product
Start free with 3 connected accounts. No sales call required, and no per-customer integration code.