Multimodal text generation with reference images, available through the OpenAI-compatible Chat Completions API with streaming.
curl --fail-with-body https://aibubu.ai/v1/chat/completions \
-H "Authorization: Bearer $AIBUBU_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $(uuidgen)" \
--data-raw '{
"stream": false,
"messages": [
{
"role": "user",
"content": "Hello"
}
],
"model": "deepseek-v4-flash-vision-exp"
}'No sample has been configured for this model. Fill in the parameters to generate.
Multimodal text generation with reference images, available through the OpenAI-compatible Chat Completions API with streaming.
A full overview is coming soon. See the model docs for parameters, or try it in the Playground tab.
Model parameter docs| Model | Capabilities | Resolution and duration | From |
|---|---|---|---|
| DeepSeek V4 Flash | Text generation | — | Input 58.5 API Credits / 1M tokens≈ $0.6429 Output 175.5 API Credits / 1M tokens≈ $1.93 |
| DeepSeek V4 Pro | Text generation | — | Input 175.5 API Credits / 1M tokens≈ $1.93 Output 526.5 API Credits / 1M tokens≈ $5.79 |