DeepSeek V4 Flash Vision Exp accepts text and image input for multimodal understanding and agent workflows. It supports the Responses API and remains compatible with Chat Completions.
Provider: DeepSeek
Category: text generation
Endpoint: /v1/responses
Status: Available
Cost: --
deepseek-v4-flash-vision-exp
DeepSeek V4 Flash Vision Exp accepts text and image input for multimodal understanding and agent workflows. It supports the Responses API and remains compatible with Chat Completions.
API access
Endpoint
POST /v1/responses
Base URL
https://api.hiapi.ai
Uses the OpenAI Responses format. One HiAPI key works across all available models.
Input price
572 Credits
/ 1M tokens
Output price
1,716 Credits
/ 1M tokens
Context
1M
context window
Max output
384K
output tokens
Try deepseek-v4-flash-vision-exp
Ask a question, upload images, or paste from clipboard. Then inspect the streamed answer, token usage, latency, and estimated cost.
Upload or paste up to 4 images, 10 MB each. Images are safety reviewed.
deepseek-v4-flash-vision-exp API Quickstart
Call deepseek-v4-flash-vision-exp with the streaming OpenAI Responses format. Copy a ready-to-use cURL, Python, or Node.js example below.
/v1/responsescurl -N -X POST "https://api.hiapi.ai/v1/responses" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-flash-vision-exp",
"input": [
{
"type": "message",
"role": "user",
"content": [
{
"type": "input_text",
"text": "分析一张图片并结合文字要求给出可执行的结论。"
}
]
}
],
"stream": true,
"store": false,
"max_output_tokens": 1024
}'Tip: Replace YOUR_API_KEY with your actual API key from the API Keys page.
Uses the OpenAI Responses format. Keep stream: true, send input as a message array, and set reasoning effort with reasoning.effort.
Combine input_text and input_image in one input message. Use a public HTTPS image URL; Playground uploads automatically provide a temporary reviewed URL.
/v1/responsescurl -N -X POST "https://api.hiapi.ai/v1/responses" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-flash-vision-exp",
"input": [
{
"type": "message",
"role": "user",
"content": [
{
"type": "input_text",
"text": "Describe the main subject, setting, and important visual details."
},
{
"type": "input_image",
"image_url": "https://static.hiapi.ai/example/subject-1.jpg",
"detail": "low"
}
]
}
],
"stream": true,
"store": false,
"max_output_tokens": 1024,
"reasoning": {
"effort": "high"
}
}'Tip: Replace YOUR_API_KEY with your actual API key from the API Keys page.
Uses the OpenAI Responses format. Keep stream: true, send input as a message array, and set reasoning effort with reasoning.effort.
Responses API
Request fields at a glance
Start with core fields, then enable advanced capabilities as needed.
Core fields
The fields that form a valid request
- model Required
- Use the model ID shown on this page.
- input Required
- Pass the current conversation context as a message array.
- stream Required
- Must be set to true.
- store Fixed
- Fixed to false and shown explicitly in examples.
Capability fields
Enable reasoning, context, and tools as needed
- reasoning.effort Optional
- medium. Examples and Playground default to medium; set this field explicitly in API requests.
- instructions Optional
- Define the role and response requirements.
- previous_response_id Conditional
- Use with compatible storage and response chaining; with store=false, replay context from the client.
- tools / tool_choice Optional
- Declare and control function tool calls.
Max output is a model specification limit, not a per-request generation target. Examples show only the runtime fields currently available and verified on HiAPI.
deepseek-v4-flash-vision-exp API Pricing
Billing follows the input, output, and cache token categories actually reported in usage. All prices below are shown per 1 million tokens.
- Input tokens
- 572 Credits/ 1M tokens
- Output tokens
- 1,716 Credits/ 1M tokens
Prompts and context sent to the model
Responses generated by the model
- Cache read
- 178.36 Credits/ 1M tokens
deepseek-v4-flash-vision-exp Model Specifications
These specifications reflect the current public HiAPI text request contract. Undisclosed capabilities are not inferred by the page.
- Streaming
- Supported
- Tool calling
- Not supported
- Structured JSON
- Supported
- Reasoning
- Not supported
- Input modalities
- text · image
- Output modalities
- text
- Reasoning efforts
- -
- Max output
- 384K
About deepseek-v4-flash-vision-exp
DeepSeek V4 Flash Vision Exp accepts text and image input for multimodal understanding and agent workflows. It supports the Responses API and remains compatible with Chat Completions.
HiAPI exposes this model through /v1/responses. Test prompts and parameters in the Playground, then use the same HiAPI API key in your server application.
- Provider
- DeepSeek
- Endpoint
- /v1/responses
- Context window
- 1M
- Released
- 2026-08-21
How to Use deepseek-v4-flash-vision-exp
The Playground and API share the same model ID, so three steps take a prompt into production.
Step 1
Validate online
Test instructions, input, and reasoning.effort in the Playground.
Step 2
Manage API keys
View an existing key or create another after signing in; one key works across all available models.
Step 3
Integrate your server
Send requests to /v1/responses and track cost with the usage object.
Frequently asked questions
Can this model generate images?
It accepts text and images and returns text. Use it for image descriptions, document and chart understanding, visual questions, and reasoning over images. Choose an image model to generate or edit images.
Which model ID and endpoint should I use?
Set model to deepseek-v4-flash-vision-exp. Use POST /v1/responses with input, or POST /v1/chat/completions with messages. Their image formats differ, so use the example for your chosen endpoint.
How do I upload images, including multiple images?
The Playground accepts uploaded or pasted JPEG, PNG, WebP and GIF files: up to 4 images per message, 10 MB each. The API accepts public HTTPS image URLs or base64 data URLs. Responses uses input_image; Chat Completions uses image_url content blocks. The stated upload count and size limits apply to the Playground.
How do I select or disable reasoning?
Responses uses reasoning.effort. The page offers none, low, medium and high, defaulting to high. none disables reasoning; other levels request reasoning effort, with latency and token use depending on the task. For Chat Completions, disable reasoning with reasoning.enabled=false.
What are the context and output limits, and why can an answer be cut short?
The context window is 1,048,576 tokens and the output budget limit is 262,144 tokens. The Playground and default examples use 1,024. Reasoning tokens count toward this budget and may consume much of it before the final answer. Raise max_output_tokens or retry without reasoning when output is truncated; combined input and output must still fit the context window.
How are images, reasoning and cached tokens billed?
Image tokens count as input, reasoning tokens count as output, and cached input is billed at the cached-input rate. See the live pricing section for all three rates. Compare usage.input_tokens, usage.output_tokens and input_tokens_details.cached_tokens against your API logs. A cache hit does not make the entire request free.
Do I need to resend conversation history?
The examples use store:false. Resend the required context in input on each turn; a previous text response alone does not retain the conversation for you. With Chat Completions, put the relevant history in messages.
Does it support JSON output and function calls?
Yes. For Responses, use text.format.type=json_object for JSON and tools/tool_choice to define and select functions. Your application executes returned function calls, then supplies a function_call_output with the matching call_id. The model does not execute local code for your application. See the API documentation for complete examples.
How do I detect the end of a streamed response?
Read text from response.output_text.delta and final usage from response.completed. Also handle response.incomplete and response.failed. Do not wait for the Chat Completions [DONE] marker on the Responses endpoint.
Can I reuse an existing HiAPI API key?
Yes, if the key has access to this model and the account has sufficient balance. Set the OpenAI SDK base_url to https://api.hiapi.ai/v1 and use responses.create or chat.completions.create. The web Playground uses your signed-in session, so you do not paste an API key there.
HiAPI Blog
