curl --request POST \
  --url https://direct.evolink.ai/v1/messages \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "deepseek-v4-flash",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Hello, world"
    }
  ]
}
'

{
  "id": "53ee6690-e14a-4e6b-890b-a135100d51c7",
  "type": "message",
  "role": "assistant",
  "model": "deepseek-v4-flash",
  "content": [
    {
      "type": "thinking",
      "thinking": "The user is asking about Japan's capital — a basic geography question. The answer is Tokyo, just give it directly.",
      "signature": "53ee6690-e14a-4e6b-890b-a135100d51c7"
    },
    {
      "type": "text",
      "text": "The capital of Japan is **Tokyo**."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 7,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "output_tokens": 77,
    "service_tier": "standard"
  }
}

Messages API

DeepSeek V4 - Anthropic-Compatible API

Invoke DeepSeek V4 models using the Anthropic Messages protocol
Supports deepseek-v4-flash / deepseek-v4-pro / deepseek-v4-flash-vision-exp
Request / response structures aligned with the Anthropic API
Image understanding: deepseek-v4-flash-vision-exp supports image content blocks (source.type may be base64 or url)
System prompts: Passed via the top-level system field
Thinking mode: Controlled by the thinking object; thinking content is returned as a content[type=thinking] block
Streaming output: SSE event stream
Tool calling: Compatible with Anthropic tool_use / tool_result flow

⚠️ Important: Sending an image block to deepseek-v4-flash / deepseek-v4-pro does not raise an error, but the upstream replaces the image with an [Unsupported Image] placeholder. The model cannot see the image, and the request is still billed. For image understanding, always set model to deepseek-v4-flash-vision-exp.

POST

messages

curl --request POST \
  --url https://direct.evolink.ai/v1/messages \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "deepseek-v4-flash",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Hello, world"
    }
  ]
}
'

{
  "id": "53ee6690-e14a-4e6b-890b-a135100d51c7",
  "type": "message",
  "role": "assistant",
  "model": "deepseek-v4-flash",
  "content": [
    {
      "type": "thinking",
      "thinking": "The user is asking about Japan's capital — a basic geography question. The answer is Tokyo, just give it directly.",
      "signature": "53ee6690-e14a-4e6b-890b-a135100d51c7"
    },
    {
      "type": "text",
      "text": "The capital of Japan is **Tokyo**."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 7,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "output_tokens": 77,
    "service_tier": "standard"
  }
}

BaseURL: The default BaseURL is https://direct.evolink.ai, which has better support for text models and long-lived connections. https://api.evolink.ai is the primary endpoint for multimodal services and serves as a fallback address for text models.

Authorizations

Authorization

string

header

required

##All APIs require Bearer Token authentication##

Get API Key:

Visit the API Key Management Page to obtain your API Key

Add to request header:

Authorization: Bearer YOUR_API_KEY

Note: Although Anthropic's native API uses the x-api-key header, EvoLink's /v1/messages uniformly uses Bearer Token authentication.

Body

application/json

model

enum<string>

default:deepseek-v4-flash

required

Model to invoke

deepseek-v4-flash: Fast general-purpose
deepseek-v4-pro: Deep reasoning
deepseek-v4-flash-vision-exp: Experimental model with image understanding; its text capabilities, parameters, and pricing match Flash

Tip: All three models have thinking enabled by default, and responses always contain a type="thinking" content block; set thinking.type="disabled" explicitly to turn it off. An unspecified or unsupported model is automatically mapped to deepseek-v4-flash.

Image understanding requires deepseek-v4-flash-vision-exp: The other two models do not raise an error for an image block, but replace it with an [Unsupported Image] placeholder and discard the actual image.

Experimental notice: -exp marks an experimental upstream release. The model name and availability may change, so implement a fallback for production use.

Available options:

deepseek-v4-flash,

deepseek-v4-pro,

deepseek-v4-flash-vision-exp

Example:

"deepseek-v4-flash"

max_tokens

integer

required

Maximum number of tokens to generate (required)

Notes:

The V4 series can reach up to 384,000
Tokens produced by thinking also count toward the max_tokens limit

Required range: 1 <= x <= 384000

Example:

1024

messages

object[]

required

List of conversation messages, alternating between user / assistant turns

Notes:

At least one message must be included
The final message is typically role=user
The image content type is supported only by deepseek-v4-flash-vision-exp; document is not supported

Minimum array length: 1

Show child attributes

system

System prompt, used to define the AI's role and behavior

Notes:

Accepts a string or an array of strings
Unlike the system message on the OpenAI endpoint, the Anthropic endpoint uses a top-level system field

Example:

"You are a helpful assistant."

temperature

number

default:1

Sampling temperature

Notes:

Range [0.0, 2.0]
Default 1; higher values are more diverse, lower values are more deterministic

Required range: 0 <= x <= 2

Example:

1

top_p

number

default:1

Nucleus sampling threshold

Notes:

Range [0, 1]
Do not adjust temperature and top_p simultaneously

Required range: 0 <= x <= 1

Example:

1

stop_sequences

string[]

Custom stop sequences

Notes:

Generation stops when the model encounters any of these strings
Up to 4 entries (following Anthropic's specification)

Maximum array length: 4

stream

boolean

default:false

Whether to return via SSE streaming

true: Server-Sent Events streaming return
false: Return the complete response all at once (default)

Example:

false

thinking

object

Thinking mode control (V4)

Notes:

Enabled by default on both models (type=enabled)
When enabled, the response content array includes a type="thinking" reasoning block (billed as output tokens)
Note: The API ignores Anthropic's native budget_tokens field; use output_config.effort to control depth
In multi-turn conversations, simply place the thinking block from the previous response back into the assistant content array verbatim (the Anthropic protocol is more lenient — missing thinking will not cause an error, but preserving the signature helps context consistency)

Show child attributes

output_config

object

Output configuration (V4 extension)

Notes: DeepSeek only supports the effort field

Show child attributes

tools

object[]

List of tool definitions

Notes:

Follows the Anthropic tool definition specification
input_schema uses a JSON Schema object

Show child attributes

tool_choice

object

Controls tool-calling behavior

Available types:

auto: Model decides automatically (default when tools are provided)
any: Must call some tool (without specifying which)
tool: Must call the specified name
none: Forbid tool calls

Show child attributes

Response

Message object

Anthropic-style message response

string

Unique message ID

type

enum<string>

Response object type

Available options:

message

role

enum<string>

Available options:

assistant

model

string

Model actually used

Example:

"deepseek-v4-pro"

content

object[]

List of response content blocks

Possible block types:

thinking: Reasoning process (only when thinking is enabled)
text: Final answer text
tool_use: Tool call initiated by the model

Show child attributes

stop_reason

enum<string>

Stop reason

end_turn: Natural completion
max_tokens: Reached the max_tokens limit
stop_sequence: Hit a stop_sequences entry
tool_use: Model triggered a tool call

Available options:

end_turn,

max_tokens,

stop_sequence,

tool_use

stop_sequence

string | null

The specific sequence hit when stop_reason=stop_sequence; otherwise null

usage

object

Token usage statistics (Anthropic specification)

Show child attributes

Image Series

Video Series

Audio Series

Text Series

Account Management

Task Management

File Management

DeepSeek V4 - Anthropic-Compatible API

Authorizations

Body

Response