DeepSeek · Fast / lightweight
DeepSeek-V4.1-Flash Updated
Native multimodal support
- Provider
- DeepSeek
- Family
- DeepSeek V4
- Model type
- Fast / lightweight
- Official release date
- Sep 10, 2026
- Release type
- New version
- Status
- Generally available
- Previous version
- DeepSeek-V4-Flash
- Input
- text, image
- Output
- text
- Context window
- 1,000,000 tokens
- Max output
- 384,000 tokens
- API model id
deepseek-flash- API price (input)
- $0.30 per 1M tokens
- API price (output)
- $1.20 per 1M tokens
- Price notes
- Peak cache-miss rates; off-peak is half ($0.15 / $0.60). Cache-hit input $0.006 peak.
- Consumer apps
- —
- Developer API
- DeepSeek API (OpenAI- and Anthropic-compatible)
- Free access
- None documented
- Paid access
- —
- Availability
- —
- Licence
- proprietary
Main improvements
- Native multimodal support
- Upgraded to snapshot 1017 on 2026-10-17 with same name and price
- Thinking and non-thinking modes
Best use cases
- Low-cost chat and agents
Benchmarks
No benchmark results recorded. We only list scores with a source, benchmark version and date.
More from DeepSeek
- DeepSeek-V4-Pro Generally available · Aug 13, 2026
- deepseek-chat and deepseek-reasoner API names deprecation Deprecated · Apr 24, 2026