# DeepSeek V4.1 Flash

DeepSeek V4.1 Flash is the smallest model in DeepSeek's new architecture family, a sparse mixture-of-experts model built for coding, reasoning, and agent workloads. It is the first DeepSeek model on the Causal Encoder-Decoder design, adds native visual understanding, and sharply cuts KV-cache requirements for faster, cheaper agent runs. Its published context window is 1,048,576 tokens. The advertised input rate is $0.30 per million input tokens. The advertised output rate is $1.20 per million output tokens.

- Model identifier: ` deepseek/deepseek-v4.1-flash `
- Creator: DeepSeek
- Providers: Azure AI Foundry, DeepSeek, Tencent, Fireworks
- Model type: Text generation
- Canonical model page: [DeepSeek V4.1 Flash](<https://impossibl.com/deepseek/deepseek-v4.1-flash>)
- API availability: Listed as serving through Impossibl.
- Creator release date: 2026-09-10

## Providers

Different companies host DeepSeek V4.1 Flash. Impossibl routes your request intelligently and handles failover automatically when a provider is down.

| Provider | Endpoint location | Routes | Verified context window | Tool calling |
| --- | --- | --- | --- | --- |
| Azure AI Foundry | East US 2 | 1 | Not verified | Not verified |
| DeepSeek | Not specified | 1 | Not verified | Not verified |
| Tencent | Not specified | 1 | Not verified | Not verified |
| Fireworks | Not specified | 1 | Not verified | Not verified |

The model price applies whichever provider serves the request. Endpoint locations do not guarantee inference residency.

## Specs and Capabilities

- Context window: 1,048,576 tokens
- Maximum output tokens: Not published in the public catalog; this is separate from the context window.
- Inputs: text, image
- Outputs: text
- Tool calling: Not published

## API access

Use ` deepseek/deepseek-v4.1-flash ` as the model identifier through the Impossibl API.

Published API endpoints:

- ` /v1/chat/completions `

[API quickstart](<https://impossibl.com/docs/agent-quickstart>) · [Models and billing documentation](<https://impossibl.com/docs/models>)

## Published API prices

All rates are in USD. An unpublished rate is unknown, not zero. Workspace-specific pricing, taxes, balance-purchase fees, billing adjustments, and tool charges can affect the amount paid.

| Usage | Published base rate | Billing unit |
| --- | --- | --- |
| Input | $0.30 | per million input tokens |
| Output | $1.20 | per million output tokens |
| Cache read | $0.006 | per million cached input tokens |
| Cache write (default retention) | Not published | per million tokens |
| Cache write (one-hour retention) | Not published | per million tokens |
| Audio input | Not published | per million audio input tokens |

## Benchmarks

Last checked: 2026-11-18T00:37:52.102Z.
Configurations are evaluated separately. API scores retain their original units. Unmeasured scores are omitted.

### DeepSeek V4.1 Flash (Max)

Source: [Evaluation source](<https://artificialanalysis.ai/models/deepseek-v4-1-flash>)

| Benchmark | Score |
| --- | --- |
| Humanity’s Last Exam | 39.25% |
| SciCode | 51.85% |
| CritPt | 14.29% |
| AA-LCR | 84.00% |
| Multilingual LCR | 23.89% |
| Harvey LAB | 1.67% |
| GDPval Elo | 1600.00 |
| GDP.pdf All-pass | 12.80% |
| AutomationBench Partial | 68.89% |
| AnalystAgent | 35.00% |
| MMMU-Pro | 76.99% |
| ITBench | 46.92% |
| legal index (API) | 41.3 |
| economics index (API) | 44.4 |
| engineering index (API) | 39.3 |
| intelligence index (API) | 39.5 |
| strategy and ops index (API) | 51.7 |
| finance and accounting index (API) | 45 |
| healthcare and medical index (API) | 40.6 |

Full benchmark data and breakdowns: [Complete benchmark snapshot](<https://api.impossibl.com/v1/models/deepseek/deepseek-v4.1-flash/benchmarks>).


## Availability and retirement

The model is listed as serving. No retirement is announced in the current public catalog.

## Frequently asked questions

### What is DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash is the smallest model in DeepSeek's new architecture family, a sparse mixture-of-experts model built for coding, reasoning, and agent workloads. It is the first DeepSeek model on the Causal Encoder-Decoder design, adds native visual understanding, and sharply cuts KV-cache requirements for faster, cheaper agent runs.

Sources: [OpenRouter model listing](<https://openrouter.ai/deepseek/deepseek-v4.1-flash>); [DeepSeek release announcement](<https://api-docs.deepseek.com/news/news260910/>)

### How much does DeepSeek V4.1 Flash cost?

DeepSeek V4.1 Flash costs $0.30 per million input tokens and $1.20 per million output tokens through the Impossibl API. Cache reads cost $0.006 per million tokens. All rates are in USD.

### What is the context length of DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash has a 1,048,576-token context window through the Impossibl API.

### What inputs and outputs does DeepSeek V4.1 Flash support?

DeepSeek V4.1 Flash accepts text and image as input. DeepSeek V4.1 Flash returns text.

### How do I use DeepSeek V4.1 Flash through an API?

DeepSeek V4.1 Flash is available through the Impossibl API. Use deepseek/deepseek-v4.1-flash as the model identifier with /v1/chat/completions. The quickstart explains how to create an API key and send your first request.

### When was DeepSeek V4.1 Flash released?

DeepSeek released DeepSeek V4.1 Flash on September 10, 2026.

Sources: [Creator release announcement](<https://api-docs.deepseek.com/news/news260910/>)

## Sources and related links

Prices, limits, capabilities, endpoints, and availability come from the public Impossibl catalog. Creator documentation supplies descriptions and release dates where available; it does not override the published API offering.

- [Public Impossibl model catalog](<https://api.impossibl.com/v1/models>)
- [Full model directory as JSON](<https://impossibl.com/models.json>)
- [DeepSeek V4.1 Flash: model details and pricing](<https://impossibl.com/deepseek/deepseek-v4.1-flash>)
- [Model directory](<https://impossibl.com/models>)
- [OpenRouter model listing](<https://openrouter.ai/deepseek/deepseek-v4.1-flash>)
- [DeepSeek release announcement](<https://api-docs.deepseek.com/news/news260910/>)
