DeepSeek V4 Flash Vision API

Name: DeepSeek V4 Flash Vision
Brand: DeepSeek
Price: 0.308 USD
Availability: InStock

DeepSeek V4 Flash Vision is a fast vision-language model (experimental) for image understanding and multimodal chat at Flash-tier cost — zero platform fee.

$0.308

Input / 1M tokens

$1.231

Output / 1M tokens

VLM

Vision + text

Platform fee

Why use DeepSeek V4 Flash Vision on NovAI?

Fast vision-language - Flash-tier multimodal understanding
Image + text prompts - OCR, charts, scenes and documents
Cost-efficient - vision at a fraction of flagship pricing
Experimental build - latest DeepSeek vision capabilities
Zero platform fee - raw provider list price, no markup
OpenAI-compatible - drop-in via /v1/chat/completions

Best use cases

Image captioning and description
Screenshot and document OCR
Chart and diagram parsing
Visual question answering
Multimodal chat assistants

Quick start

cURL

curl https://aiapi-pro.com/v1/chat/completions \
  -H "Authorization: Bearer $NOVAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash-vision",
    "messages": [{"role":"user","content":"Describe what you see in this image"}]
  }'

Python (OpenAI SDK)

from openai import OpenAI
client = OpenAI(
    base_url="https://aiapi-pro.com/v1",
    api_key="YOUR_NOVAI_API_KEY",
)
resp = client.chat.completions.create(
    model="deepseek-v4-flash-vision",
    messages=[{"role":"user","content":"Describe what you see in this image"}],
)
print(resp.choices[0].message.content)