DeepSeek V4 Flash
Fast and lightToolsDeepSeek · Released April 2026
DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...
It is part of DeepSeek's open-weight line, known for delivering near-frontier reasoning and coding quality at a fraction of typical frontier pricing.
Facts and pricing
- Provider
- DeepSeek
- Context window
- 1.0M tokens
- Input price
- 0.99 credits / 1M tokens ($0.090 raw)
- Output price
- 2.0 credits / 1M tokens ($0.18 raw)
- Vision (image input)
- No
- Tool calling
- Yes
- Extended reasoning
- No
Credits are what Zeplik bills: 1 credit = $0.10, computed from the raw provider rate with a 1.10x margin. Raw prices shown per 1M tokens.
Try DeepSeek V4 Flash now
What DeepSeek V4 Flash is best for
- High-volume everyday tasks where speed and cost matter: summaries, drafts, quick questions
- Very long documents and codebases: a 1.0M-token window fits entire books or repositories in one conversation
- Tool use and agents: reliably calls functions, so it can search, run skills and drive workflows
Example prompts
Prompts that suit a fast and light model like DeepSeek V4 Flash:
- Summarize this article in five bullet points
- Rewrite this email to be shorter and friendlier
- Give me ten name ideas for a hiking newsletter
- Translate this paragraph to Spanish and keep the tone
DeepSeek family
| Model | Context | Input cr/M | Output cr/M | Released |
|---|---|---|---|---|
| DeepSeek V4 Pro | 1.0M | 4.8 | 9.6 | April 2026 |
| DeepSeek V4 Flash | 1.0M | 0.99 | 2.0 | April 2026 |
| DeepSeek V3.2 | 131K | 2.5 | 3.8 | December 2025 |
| DeepSeek V3.2 Exp | 164K | 3.0 | 4.5 | September 2025 |
| DeepSeek V3.1 Terminus | 164K | 3.0 | 10.5 | September 2025 |
| DeepSeek V3.1 | 164K | 2.3 | 8.7 | August 2025 |
| R1 0528 | 164K | 5.5 | 23.7 | May 2025 |
| DeepSeek V3 0324 | 164K | 2.6 | 9.9 | March 2025 |
| R1 Distill Llama 70B | 128K | 8.8 | 8.8 | January 2025 |
| R1 | 164K | 7.7 | 27.5 | January 2025 |
Compare DeepSeek V4 Flash
- DeepSeek V4 Flash vs GPT-5.6 Luna
- DeepSeek V4 Flash vs Gemini 3.5 Flash
- DeepSeek V4 Flash vs DeepSeek V4 Pro
Frequently asked questions
- What is DeepSeek V4 Flash?
- DeepSeek V4 Flash is an AI model by DeepSeek. DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...
- How much does DeepSeek V4 Flash cost on Zeplik?
- Input tokens cost 0.99 credits per million and output tokens 2.0 credits per million (1 credit = $0.10; the raw provider rates are $0.090 and $0.18 per million). New accounts start with free credits.
- How long can a conversation with DeepSeek V4 Flash be?
- DeepSeek V4 Flash has a 1.0M-token context window (1,048,576 tokens), which covers the conversation plus any documents you attach.
- Does DeepSeek V4 Flash support images and tools?
- DeepSeek V4 Flash is text-only and supports tool calling.
Related models
DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,...
DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...
DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...
DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further optimizing the model's...
DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context...
May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active...