Model Overview
DeepSeek-V3 (deepseek-chat) is a powerful large language model developed by DeepSeek AI, designed for high-quality conversational AI applications with strong reasoning capabilities.
Key Features
- High intelligence (3/4 dots rating)
- Fast speed (4/5 lightning bolts rating)
- 64,000 context window
- 8,192 max output tokens
- 2023 knowledge cutoff (estimated)
- Text input support
- Text output support
Technical Specifications
- Pricing:
- Standard hours (08:30-00:30 Beijing Time):
- $0.069 per 1M tokens (input with cache hit)
- $0.276 per 1M tokens (input without cache hit)
- $1.102 per 1M tokens (output)
- Discount hours (00:30-08:30 Beijing Time):
- $0.034 per 1M tokens (input with cache hit)
- $0.138 per 1M tokens (input without cache hit)
- $0.551 per 1M tokens (output)
- Supports: Input: text; Output: text only
- Features: Context caching, optimized for chat applications
Snapshots
- deepseek-chat (current version)
- deepseek-v3
Positioning and Use Cases
DeepSeek-V3 is positioned as a high-performance general-purpose language model with excellent reasoning capabilities. It excels at complex conversations, code generation, and creative writing tasks. The model is particularly well-suited for applications requiring deep contextual understanding due to its large 64K token context window. With its competitive pricing structure and discount hours, it offers an excellent balance of performance and cost-effectiveness for developers and businesses building AI applications.
Additional Notes
- Default maximum output length is 4K tokens if not specified by the user
- Context caching feature helps reduce costs for repeated or similar queries
- When both recharged balance and bonus balance exist, bonus balance will be deducted first
- The model supports efficient processing of long documents and conversations with its large context window
Documentation
Official Documentation
