DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name. V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on output from a 552B-parameter backbone, an asymmetric split that keeps per-token compute low relative to the model's total size. Image understanding is native to the architecture, with visual and text embeddings trained jointly from the start of pre-training rather than added afterward as in the earlier experimental V4 Flash Vision ExpOpens in new tab.
It is suited for coding, terminal, and computer-use agents, along with long-horizon tasks that must run to completion across many steps and long-context analysis. Compressed KV caching cuts cache memory to roughly a quarter of the previous Flash generation, significantly reducing costs on agentic workloads. DeepSeek positions it as the cost-efficient tier of the V4.1 family and reports that it exceeds V4 ProOpens in new tab on performance, speed, and task completion time.
提供商
不同公司托管同一模型。OpenRouter 根据您选择的路由模式将请求路由到其中一家公司:Balanced(价格 + 速度)、Nitro(速度最快)或 Floor(价格最低)。
| US$0.036 | US$0.15 | US$0.003 | -- | -- | ||
| US$0.050 | US$0.15 | US$0.004 | -- | -- | ||
| US$0.150 | US$0.60 | US$0.003 | -- | -- |
- 输入 US$0.036 / 输出 US$0.15-- tok/s-- 正常运行时间
- 输入 US$0.05 / 输出 US$0.15-- tok/s-- 正常运行时间
- 输入 US$0.15 / 输出 US$0.60-- tok/s-- 正常运行时间
定价
客户为此模型实际支付的平均价格,与提供商公布的价格并列展示。由于缓存和折扣,实际支付的价格往往远低于标价。
DeepSeek V4.1 Flash 1017 (batch) — 价格历史
| US$0.036 | US$0.15 | US$0.003 | — | ||
| US$0.050 | US$0.15 | US$0.004 | — | ||
| US$0.150 | US$0.60 | US$0.003 | — |
- 输入 US$0.036 / 输出 US$0.15
- 输入 US$0.050 / 输出 US$0.15
- 输入 US$0.150 / 输出 US$0.60
性能
吞吐量是模型生成内容的速度(每秒 token 数,越高越好)。延迟是完整的往返时间(越低越好)。TTFT 是首个 token 的响应时间,即您看到内容出现前需要等待的时间(越低越好)。
正常运行时间
正常运行时间是指过去 3 天中至少有一家提供商响应请求的时间百分比。可用性是指成功提供推理服务的时间百分比。OpenRouter 会持续监控,并在某家提供商返回错误时使用次优提供商。
正常运行时间数据尚不足,暂时无法显示。
当上游提供商发生错误时,如果您的请求筛选条件允许,我们可以通过路由到另一家健康的提供商来恢复服务。您可以通过以下接口以编程方式访问各提供商的正常运行时间数据: 端点 API。了解更多有关我们的 负载均衡和提供商路由选项。
基准测试
标准化评测得分,以及此模型针对共用基准测试提示词的实际输出。百分比越高越好,每项排名显示此模型在该来源所评测的模型中的位置。
| 来源 | 基准测试 | 得分 |
|---|---|---|
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Intelligence Index | 28.6 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Finance & Accounting Index | 39.5 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Strategy & Ops Index | 48.4 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Legal Index | 33.9 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Engineering Index | 34.4 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Economics Index | 35.3 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Openness Index | 44.4 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) AutomationBench-AA | 68.5% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) GDP.pdf | 14.2% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) HLE-Diamond | 20.8% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Terminal-Bench 4.0 | 28.8% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) FrontierCode-AA 2.0 | 16.8% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Terminal-Bench-Science | 10.5% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) CritPt v2 (Hard Subset) | 27.5% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) CritPt v2 | 52.3% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) AA-Omniscience Accuracy | 38.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) AA-Omniscience Non-Hallucination Rate | 38.8% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) MMMU-Pro | 79.7% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Intelligence Index | 32.6 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Finance & Accounting Index | 42.2 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Strategy & Ops Index | 49.2 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Legal Index | 37.7 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Healthcare & Medical Index | 41.3 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Engineering Index | 37.6 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Economics Index | 38.5 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Cyber Index | 56.8 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Openness Index | 44.4 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) AutomationBench-AA | 66.6% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) GDP.pdf | 17.2% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) HLE | 43.3% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) AA-LCR | 82.7% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) GDPval-AA | 62.3% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) CritPt | 24.0% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) HLE-Diamond | 25.3% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) SciCode | 55.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Terminal-Bench 4.0 | 33.8% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) FrontierCode-AA 2.0 | 28.5% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) FrontierCode 1.1 | 43.7% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) Terminal-Bench-Science | 12.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) CritPt v2 (Hard Subset) | 25.0% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) CritPt v2 | 50.6% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) AA-Omniscience Accuracy | 39.6% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) AA-Omniscience Non-Hallucination Rate | 35.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) CWE-Bench-AA | 69.2% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) DeepsecBench-AA | 24.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) CyberGym-E2E-AA | 76.3% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) MLCR-AA | 26.7% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) MMMU-Pro | 80.6% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Max) VIALS-AA | 40.4% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Intelligence Index | 18.5 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Finance & Accounting Index | 37.2 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Strategy & Ops Index | 39.7 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Legal Index | 31.7 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Engineering Index | 25.0 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Economics Index | 32.1 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Openness Index | 44.4 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) AutomationBench-AA | 48.8% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) GDP.pdf | 4.4% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) HLE | 11.5% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) AA-LCR | 57.3% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) GDPval-AA | 40.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) CritPt | 0.3% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) HLE-Diamond | 7.1% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) SciCode | 38.5% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Terminal-Bench 4.0 | 9.1% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) FrontierCode-AA 2.0 | 6.1% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Terminal-Bench-Science | 2.4% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) CritPt v2 (Hard Subset) | 1.0% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) CritPt v2 | 4.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) AA-Omniscience Accuracy | 21.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) AA-Omniscience Non-Hallucination Rate | 74.7% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) MMMU-Pro | 69.2% |
| Design Arena | DeepSeek-V4.1-Flash (1017) Agents Arena Spreadsheet Elo | 1265 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena 3D Elo | 1337 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Asciiart Elo | 1221 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Code Categories Elo | 1279 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Data Visualization Elo | 1270 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Game Development Elo | 1248 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena SVG Elo | 1234 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena UI Component Elo | 1278 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Website Elo | 1290 |
| Vals AI | DeepSeek V4.1 Flash 1017 Access to Justice Benchmark | 73.1% |
| Vals AI | DeepSeek V4.1 Flash 1017 Code Migration v1.1 | 67.5% |
| Vals AI | DeepSeek V4.1 Flash 1017 CyberOffense | 54.8% |
| Vals AI | DeepSeek V4.1 Flash 1017 EMB v1.1 | 69.2% |
| Vals AI | DeepSeek V4.1 Flash 1017 Finance Agent v2.1 | 62.9% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vals Finance Index | 60.9% |
| Vals AI | DeepSeek V4.1 Flash 1017 HLAB-Vals-1.1 | 14.2% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vals Legal Index | 71.9% |
| Vals AI | DeepSeek V4.1 Flash 1017 Legal Research Bench | 45.2% |
| Vals AI | DeepSeek V4.1 Flash 1017 Robo-bench | 4.2% |
| Vals AI | DeepSeek V4.1 Flash 1017 Tax Agent Bench | 44.4% |
| Vals AI | DeepSeek V4.1 Flash 1017 Terminal-Bench 4.0 | 31.8% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vals Index | 61.3% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vibe Code Bench 1-100 v1.1 | 26.2% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vibe Code Bench v1.1 | 87.7% |
应用
向此模型发送最多流量的公开应用。这能很好地反映实际生产工作负载的情况,也能提示此模型最适合哪些用例。
使用此模型的应用尚无足够数据。
活动
此模型的 token 用量和请求流量随时间的变化。
API
可直接用于调用此模型的代码。OpenRouter 的 API 与 OpenAI 兼容——大多数 SDK 只需替换基础 URL 即可使用。切换模型时,唯一需要更改的是下方的模型标识符。
目前无法通过 EU 区域内路由使用。 区域内路由文档
文档
在 OpenRouter 上使用 DeepSeek: DeepSeek V4.1 Flash 1017 (batch) 构建应用的指南、实用教程、博客文章和工具。
常见问题
DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name. V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.
OpenRouter 上的 DeepSeek V4.1 Flash 1017 (batch) 文档包括 Best OCR Model for Scanned Documents, Measured 博客文章、Build an AI Code Reviewer with OpenRouter 博客文章、Open-Weight vs Closed-Weight Models: What's the Difference? 博客文章、How to Extract Tables from a PDF with an LLM 博客文章和Does DeepSeek V4 Have Vision? 博客文章。快速入门介绍了 API 密钥和您的首次请求,API 参考列出了所有参数。
DeepSeek V4.1 Flash 1017 (batch) 的费用为 US$0.036/百万输入 token 和 US$0.15/百万输出 token,Cache Read,价格为 US$0.003/M tokens 单独计费。
DeepSeek V4.1 Flash 1017 (batch) 的上下文窗口为 1,048,576 个 token。它最多支持 1,048,576 个补全 token。
支持。在提供 DeepSeek V4.1 Flash 1017 (batch) 的 3 家提供商中,有 2 家支持该模型通过 tools 和 tool_choice 进行函数调用,包含工具的请求会被路由到这些提供商。 它还支持通过 response_format 中的 JSON schema 实现结构化输出。
DeepSeek V4.1 Flash 1017 (batch) 接受 文本和图像 作为输入,并返回 文本。
在 OpenRouter 上,DeepSeek V4.1 Flash 1017 (batch) 由 3 家提供商提供服务:Morph、Sail Research和Together。 请求会路由至最佳的可用提供商,并在故障时自动切换至其他提供商,您也可以通过提供商路由固定或排除提供商。
DeepSeek V4.1 Flash 1017 (batch) 于 2026年10月17日 发布。
