DeepSeek V4.1 Flash
DeepSeek V4.1 Flash is a large language model made by DeepSeek, added to the catalog on 10 Sep 2026. It has a context window of 1,048,576 tokens and costs $0.0498 per 1M input tokens and $0.199 per 1M output tokens. It accepts text and images and produces text. Typical uses: summarising lengthy documents, working with text and image inputs and extracting information from supplied material.
Price checked ยท source: OpenRouter
What DeepSeek V4.1 Flash is
DeepSeek V4.1 Flash is a model from DeepSeek that accepts text and image inputs and produces text. It has a context window of 1,048,576 tokens and a maximum output of 393,216 tokens. The context limit provides room for lengthy prompts and reference material, while the output limit supports extended responses. The provider describes its architecture as a sparse mixture of experts and identifies it as the first model built on its Causal Encoder-Decoder architecture.
The model supports reasoning, tool use, structured outputs, and response formats. These options make it suitable for tasks that need a response organized to a requested format or that involve tools in the interaction. Since it accepts images as well as text and returns text, it can be considered for workflows combining visual inputs with written context and text-based results. Its long context is relevant to processing large collections of notes, lengthy documents, or multiple supplied materials in one prompt, including material from different sources. Example work includes summarizing material, extracting or organizing information, drafting content from supplied context, and working through multi-step questions. The catalog lists capabilities rather than specific benchmark results or domain guarantees, so selection should depend on whether these inputs, outputs, and controls fit the intended workflow.
Key features
- Accepts text and image inputs and produces text
- Has a context window of 1,048,576 tokens
- Supports tool use, reasoning, structured outputs, and response formats
- Supports a maximum output of 393,216 tokens
Typical uses
- Summarising lengthy documents
- Working with text and image inputs
- Extracting information from supplied material
- Producing structured text responses
- Answering multi-step questions
DeepSeek V4.1 Flash pricing
US dollars, list price through OpenRouter.
| Item | Standard | Batch |
|---|---|---|
| Input, per 1M tokens | $0.0498 | $0.05 |
| Output, per 1M tokens | $0.199 | $0.15 |
| Cached input read, per 1M tokens | $0.001 | $0.004 |
| Variant | Model id |
|---|---|
| Batch | deepseek/deepseek-v4.1-flash:batch |
A variant is the same model under another id, with prices or limits of its own. Its prices are the columns of the list above.
The price last changed on 21 Nov 2026. 3 price changes have been recorded since DeepSeek V4.1 Flash was added.
| Date | Input /1M | Output /1M | Change |
|---|---|---|---|
| $0.0498 | $0.199 | Price lowered | |
| $0.05 | $0.80 | Price changed | |
| $0.0507 | $0.203 | Price lowered | |
| $0.30 | $1.20 | Added at this price |
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0001 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0014 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0019 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.0078 |
DeepSeek V4.1 Flash specifications
| Model id | deepseek/deepseek-v4.1-flash |
|---|---|
| Version | deepseek-v4.1-flash-20260910 |
| Context window | 1,048,576 tokensAbout 786,432 English words |
| Max output | 393,216 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 10 Sep 2026 |
| Input types | text, image |
|---|---|
| Output types | text |
| Tool calling | Yes |
| Structured outputs | Yes |
| Reasoning effort control | Yes |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for DeepSeek V4.1 Flash yet.
Compare DeepSeek V4.1 Flash
Side by side with other models: prices, context window, what each accepts and supports, and the same four tasks priced on both.