deepseek-v4-flash · DeepSeek · text
| Off-Peak Hours | |
|---|---|
| Input | $0.22/M |
| Output | $0.67/M |
| Cache Read | $0.0075/M |
Peak hours: 09:00-12:00 · 14:00-18:00
DeepSeek V4 Flash is DeepSeek's ultra-fast model, offering 1M context with an ultra-low deduction coefficient (0.5) and excellent value. Ideal for high-concurrency, low-latency scenarios: customer support, classification, summarization, and lightweight code completion. Supports disk caching.
1M context; disk caching (cache_hit ~14% of cache_miss); visual understanding; Function Calling; JSON mode; FIM completion; OpenAI API compatible; fast responses.
High-concurrency customer support; classification and tagging; summarization; lightweight code completion; Chinese-language understanding; batch data processing.
Does not generate images; weaker reasoning than Pro; caching depends on hits (not guaranteed); max_tokens capped at 8K.
Anthropic official CLI coding agent. Multi-file edits, tool use, long-context reasoning.
export ANTHROPIC_AUTH_TOKEN="YOUR_API_KEY"
export ANTHROPIC_BASE_URL="https://api.robovai.com"
export ANTHROPIC_MODEL="deepseek-v4-flash"
claude