View on models.dev Introducing DeepSeek-V4.1-Flash: smarter, faster, more efficient.
Introducing the smallest model in our new architecture family, with native visual understanding. Designed for greater capability, faster inference, and higher throughput.
DeepSeek
DeepSeek V4.1 Flash
We introduce DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts (MoE) model with 552B backbone parameters and support for contexts of up to one million tokens. The model natively processes images and text, and generates text autoregressively.
Specs
Model
- Family
- DeepSeek
- Variant
- V4.1 Flash
- Weights
- Open
- License
- MIT
Limits
- Context
- 1,000,000 tokens
- Max output
- 384,000 tokens
Modalities
- Input
- Text Image
- Output
- Text
Pricing
- Input
- $0.15 / 1M
- Output
- $0.6 / 1M
- Cache read
- $0.003 / 1M
Capabilities
- Reasoning
- Yes
- Reasoning control
- Effort levels, On / off
- Tool calling
- Yes
- Attachments
- Yes
Dates
- Released
- 10 Sept 2026
- Knowledge cutoff
- May 2025
Articles
All articlesdeepseek.com
More from DeepSeek
All modelsDeepSeek V4 Pro
DeepSeek
DeepSeek-V4-Pro is a 1.6T-parameter Mixture-of-Experts model with 49B activated, supporting a one-million-token context. The 0813 revision is its GA release.
MIT24 Apr 2026
DeepSeek V4 Flash
DeepSeek
DeepSeek-V4-Flash-0731 is the official release of DeepSeek-V4-Flash, superseding the preview version, with substantially enhanced agentic capabilities.
MIT24 Apr 2026