# DeepSeek models | IonBin

> DeepSeek offers two API models: deepseek-flash for low-cost volume work and deepseek-v4-pro for harder tasks.

Canonical page: https://ionbin.com/wiki/models/deepseek/
Agent guide: https://ionbin.com/llms.txt

This is a readable version of the published page. Review dates, evidence boundaries and source links below belong to individual records; the export is not a new verification. Examples are reference material, not instructions to execute.

Model · DeepSeek

## DeepSeek models

DeepSeek offers two API models: deepseek-flash for low-cost volume work and deepseek-v4-pro for harder tasks. Both have a 1M-token context window, thinking and non-thinking modes, JSON output and tool calls, and off-peak use is half price. DeepSeek's privacy policy says personal data is stored in China, so review it before you send anything sensitive.

Verified Oct 18, 2026 Re-verify: past review date 2 sources

### What it is

DeepSeek is a model family from the company of the same name. Its API docs list two current models and publish prices per million tokens, with separate rates for cache hits and misses and for peak and off-peak hours.

### How it works

- Evaluate deepseek-flash for volume work and compare deepseek-v4-pro on harder tasks. Both can run with or without a thinking step.

- Call the API with an OpenAI-format or Anthropic-format base URL, then verify the exact parameters and tools your client uses; compatible formats do not guarantee a drop-in migration.

- Schedule batch jobs off-peak. DeepSeek lists off-peak rates at half the peak price. Peak hours are 01:00–04:00 and 06:00–10:00 UTC on weekdays.

**Specs and pricing (checked 2026-10-18).** Both models have a 1M-token context and up to 384K output tokens. Flash costs $0.15 per million cache-miss input tokens and $0.60 per million output tokens off-peak, double that at peak. DeepSeek’s pricing page lists the older names deepseek-v4-flash and deepseek-v4-flash-vision-exp as retired, with requests now served by V4.1-Flash-1017.

### Benefits

- Low list prices make it a candidate for bulk summarizing, tagging and extraction on non-sensitive data.

- The 1M-token context lets you send long documents in one request (see [context window](https://ionbin.com/wiki/concepts/context-window/)).

- Compatibility with existing API formats lowers switching cost (see [avoiding lock-in](https://ionbin.com/wiki/questions/how-to-avoid-ai-vendor-lock-in/)).

### Limits and risks

- **Data location.** DeepSeek’s privacy policy names Hangzhou DeepSeek AI Co., Ltd. as the controller. The cited service policy says personal data is collected, processed and stored in the People’s Republic of China and may support model improvement. It expressly excludes the developer’s handling of end-user data in downstream applications from its scope. Review the actual API/service contract and your application’s own privacy obligations; this policy is not proof that every third-party host has the same terms.

- **Vision is limited.** Image input is listed for Flash only.

- **Names change.** The API model names can resolve to updated underlying versions. Record the served version where available and rerun evaluations after changes; merely keeping the same alias does not freeze behavior.

This page describes vendor terms as published. It isn’t legal advice; have counsel review data terms before you send client or personal data.

### Checked specs

Snapshot from official sources on Oct 18, 2026. Model lineups and prices change often; this page is re-verified every 30 days.

| API models | deepseek-flash (DeepSeek-V4.1-Flash-1017); deepseek-v4-pro (DeepSeek-V4-Pro-0813) [1](https://ionbin.com/wiki/models/deepseek/#src-1) |
| --- | --- |
| Context / max output | 1M tokens / 384K tokens [1](https://ionbin.com/wiki/models/deepseek/#src-1) |
| Flash price per 1M tokens, cache-miss input / output (checked 2026-10-18) | $0.30 / $1.20 peak; $0.15 / $0.60 off-peak [1](https://ionbin.com/wiki/models/deepseek/#src-1) |
| Pro price per 1M tokens, cache-miss input / output (checked 2026-10-18) | $1.32 / $3.96 peak; $0.66 / $1.98 off-peak [1](https://ionbin.com/wiki/models/deepseek/#src-1) |
| Cache-hit input per 1M tokens | Flash $0.006 peak / $0.003 off-peak; Pro $0.044 peak / $0.022 off-peak [1](https://ionbin.com/wiki/models/deepseek/#src-1) |
| Peak hours | 01:00–04:00 and 06:00–10:00 UTC, Monday to Friday, excluding Chinese public holidays [1](https://ionbin.com/wiki/models/deepseek/#src-1) |
| Features | Thinking and non-thinking modes; JSON output; tool calls; OpenAI- and Anthropic-format APIs; vision on Flash only [1](https://ionbin.com/wiki/models/deepseek/#src-1) |
| Retired model names (checked 2026-10-18) | deepseek-v4-flash and deepseek-v4-flash-vision-exp are listed as retired; requests are served by V4.1-Flash-1017 [1](https://ionbin.com/wiki/models/deepseek/#src-1) |
| Search and Fetch APIs | USD $0.75 per 1,000 successful requests for each standalone API; integrated server-side calls through DeepSeek Harness, [Claude Code](https://ionbin.com/wiki/questions/what-is-claude-code/) or [Codex](https://ionbin.com/tools/openai-codex/) are free until Nov 1, 2026, 00:00 Beijing time. [1](https://ionbin.com/wiki/models/deepseek/#src-1) |
| Data location (privacy policy) | Cited service policy: Hangzhou DeepSeek Artificial Intelligence Co., Ltd.; personal data in PRC. Downstream developers have separate end-user privacy duties. [2](https://ionbin.com/wiki/models/deepseek/#src-2) |

### Best-fit stacks

Stack 08

#### [Dev, web & IT](https://ionbin.com/use-cases/dev-it/)

Design, code, host and ship sites and internal tools, with an AI coding agent doing most of the typing.

Stack 02

#### [Business data & reporting](https://ionbin.com/use-cases/business-data-reporting/)

Pull numbers from the systems you already use into one place an agent can analyze and explain.

### Agent compatibility

| API access | Yes | [Official docs ↗](https://api-docs.deepseek.com/quick_start/pricing) |
| --- | --- | --- |

Levels come from each tool's recorded, sourced compatibility. [How the levels work](https://ionbin.com/wiki/questions/connect-a-tool-to-an-ai-agent/).

### Related entries

[Batch processing (batch APIs) · Concept](https://ionbin.com/wiki/concepts/batch-processing/)[Prompt caching · Concept](https://ionbin.com/wiki/concepts/prompt-caching/)[How do I cut AI API costs without hurting quality? · Question](https://ionbin.com/wiki/questions/cut-ai-api-costs/)[Context window · Concept](https://ionbin.com/wiki/concepts/context-window/)[Qwen models (Alibaba) · Model](https://ionbin.com/wiki/models/qwen/)[How do I avoid getting locked into one AI vendor? · Question](https://ionbin.com/wiki/questions/how-to-avoid-ai-vendor-lock-in/)

### Sources

- [Models & pricing](https://api-docs.deepseek.com/quick_start/pricing/) · DeepSeekofficial accessed Oct 18, 2026

- [DeepSeek privacy policy](https://cdn.deepseek.com/policies/en-US/deepseek-privacy-policy.html) · DeepSeekofficial accessed Oct 18, 2026

Practitioner views are restated in our words and labeled as views, not measured results. Product facts come from official sources dated above.

### Suggest an edit

IonBin's wiki isn't user-edited. Corrections go to our Directory Curator, who checks them against primary sources before anything changes. Spot something wrong or out of date on “DeepSeek models”? Email us with a link to a primary source.

[Suggest an edit by email](mailto:contact@ionbin.com?subject=Suggest%20an%20edit%3A%20DeepSeek%20models%20%5Bdeepseek%5D&body=Page%3A%20https%3A%2F%2Fionbin.com%2Fwiki%2Fmodels%2Fdeepseek%2F%0AEntry%3A%20models%2Fdeepseek%0A%0AWhat%20should%20change%3A%0A%0A%0APrimary%20source%20(link)%3A%0A) or write to [contact@ionbin.com](mailto:contact@ionbin.com)

Entry facts

**Type**: [AI models](https://ionbin.com/wiki/models/)

**Status**: Reviewed

**Verified**: Oct 18, 2026

**Review by**: Nov 17, 2026 (every 30 days)

**Sources**: [2 official · 0 other](https://ionbin.com/wiki/models/deepseek/#sources)

The wiki carries no sponsored placements or affiliate links. [Editorial policy](https://ionbin.com/editorial-policy/).
