Choose a model for the work you need to doMe
Model specifications and sources
The official DeepSeek API model ID deepseek-flash currently serves DeepSeek-V4.1-Flash. Its official specification table lists a 1M context window, up to 384K output tokens, and image understanding. These are published specifications, not results tested on this site.
- Context window
- 1,000,000 tokens
- Maximum input
- Not published / unverified
- Maximum output
- 384,000 tokens
- Input modalities
- Text, Image
- Output modalities
- Text
- Default snapshot
DeepSeek-V4.1-Flash- Documentation checked
- 2026-09-24
- Automatic model identity check
- Exact model ID confirmed · 2026-11-11
- Identity check source
- Official evidence page ↗
- Documentation source
- View source ↗
Official APIChinaVersion 1
Writing, coding, and reasoning through the original chat API.
Writing & chatCoding
API BASE URL
https://api.deepseek.comActual API model ID
deepseek-flashRequest method / path
POST /chat/completionsAuthentication
Authorization: Bearer $API_KEYEdit request parameters
Edit JSON within the verified parameter ranges. Copy and export stop if types, required fields, or enum values are invalid.
Related cloud services and applications
All services →| Service | Use case | Features and conditions | Product entry |
|---|---|---|---|
| UCloud AstraFlowMulti-model API platform | Multi-model API integration and switching | UCloud's multi-model service. Its product page describes access to 200+ models with a shared API key and OpenAI-compatible integration. Follow the platform's registration and identity verification process. Model versions, regions, credits and prices depend on its catalog and your account. | View product ↗Register ↗ |