API Hop

Provider family

DeepSeek API pricing and model selection

Compare current DeepSeek API models, API Hop token prices, and the shared Chat Completions contract.

Compare Flash and Pro model IDs and token prices, then use the shared Chat Completions endpoint. Start with the Flash request below and evaluate both models with your own prompts.

Public contract

Interfaces and completion behavior

Check the endpoint and completion behavior before choosing a client. Supported interfaces can differ between models from the same provider.

Chat Completions POST /v1/chat-completions

Change the model field between DeepSeek family IDs; authentication and endpoint remain the same.

Current catalog facts

Model IDs and pricing

API Hop catalog prices are the customer billing source. Provider list prices are dated reference data, not a savings promise.

llm

DeepSeek V4 Flash

deepseek-v4-flash
active

Use DeepSeek V4 Flash for text generation.

Provider
DeepSeek
Billing unit
1M tokens
API formats & endpoints
Supported inputs
Text
Published capabilities
Text generation

API Hop price

Input
$0.3429 /M
Output
$1.0286 /M
Cache input
$0.0114 /M

llm

DeepSeek V4 Pro

deepseek-v4-pro
active

Use DeepSeek V4 Pro for text generation.

Provider
DeepSeek
Billing unit
1M tokens
API formats & endpoints
Supported inputs
Text
Published capabilities
Text generation

API Hop price

Input
$1.0286 /M
Output
$3.0857 /M
Cache input
$0.0343 /M

Documented facts

Know the contract before the first request

Non-streaming
One complete Chat Completions JSON response
Streaming
SSE when stream is true, ending with [DONE]
Billing
Input, cached input, cache write, and output token units when configured
Selection
Flash is the routine evaluation starting point; Pro prioritizes complex reasoning evaluations

Integration limits

Check before sending a request

  • The positioning labels are not latency, throughput, or quality guarantees.
  • DeepSeek models on this page are not claimed to support Responses or Claude Messages.
  • Direct DeepSeek pricing and serving versions are separate from the API Hop service contract.

Supported workflow

What you can build from this contract

01

One endpoint for currently cataloged DeepSeek model IDs

02

Current API Hop input, cached input, and output token prices

03

Direct paths into the exact Flash model page and current API Catalog

Runnable request

Make your first request

POST /v1/chat-completions
curl -X POST "https://api.apihop.ai/v1/chat-completions" \
  -H "Authorization: Bearer <API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"Create a reversible migration checklist."}],"stream":false}'

Use a placeholder in source control. Never put an API key in a URL, analytics payload, screenshot, or committed file.

Continue the workflow

Related APIs and guides