Hacker News Best

DeepSeek-v4.1-Flash-1017

6.7内容质量
DeepSeek-v4.1-Flash-1017

TL;DR · AI 摘要

Title: deepseek-ai/DeepSeek-V4.1-Flash-1017 · Hugging Face URL Source: Markdown Content: Technical Report 👁️ Introductio...

核心要点

  • 主题聚焦：DeepSeek-v4.1-Flash-1017
  • 来源：Hacker News Best，建议结合原文判断细节。
  • AI 分析暂不可用，本条为保底评分与摘要。
#AI#编程#后端#安全#产品
打开原文
Image 1: DeepSeek-V4.1
Image 1: DeepSeek-V4.1
  • * *

![Image 2: Homepage](https://www.deepseek.com/)![Image 3: Chat](https://chat.deepseek.com/)

![Image 4: Hugging Face](https://huggingface.co/deepseek-ai)![Image 5: Twitter Follow](https://twitter.com/deepseek_ai)

Image 6: License
Image 6: License

**Technical Report** 👁️

Introduction

DeepSeek-V4.1-Flash-1017 is a minor update to **DeepSeek-V4.1-Flash**, delivering substantial improvements across reasoning, coding, agentic tasks, and visual benchmarks. Alongside these gains, the model reduces task execution time and output length, enabling users to complete tasks faster and at lower cost and advancing the performance–cost Pareto frontier across multiple agent harnesses.

| Benchmark | DS-V4.1-Flash-1017 | DS-V4.1-Flash | K3 | GLM-5.3 | GPT-5.6 Sol | Opus 5 | | --- | --- | --- | --- | --- | --- | --- | | HLE w/o tools | 42.6 (44.0†) | 36.8 (39.1†) | 43.5 | 42.0† | 44.5 | 56.3 | | Terminal-Bench 4.0 | 42.3 | 31.2 | 12.6 | 37.9 | 39.9 | 51.8 | | DeepSWE v1.1 | 77.0 | 74.2 | 67.5 | 66.9 | 73.0 | 74.0 | | ProgramBench | 29.1 | 20.3 | 17.5 | 19.0 | 23.0 | 37.0 | | NL2Repo-Bench | 68.2 | 65.4 | 58.0 | 58.0 | 56.8 | 75.3 | | Automation-Bench v1.0.6 | 60.5 | 54.8 | 46.7 | 48.8 | 45.8 | 50.3 | | Agents’ Last Exam | 35.4 | 31.8 | 27.6 | 28.5 | 26.7 | 28.6 | | Apex Bench v1.1 | 72.1 | 64.6 | 47.8 | 56.6 | 54.3 | 65.8 | | Chartography (w/tools) | 83.0 | 78.9 | 68.1 | — | 79.9 | 84.0 | | BabyVision (w/tools) | 92.5 | 89.6 | 85.7 | — | 88.9 | 94.1 | | ZeroBench-main (w/tools) | 51.0 | 49.0 | 41.0 | — | 53.0 | 52.0 |

_† denotes the text-only subset of HLE. We use the Claude Code harness for Terminal-Bench 4.0 and the Mini-SWE-V2 harness for DeepSWE v1.1._

Prompt Encoding

This release does not include a Jinja-format chat template. The `encoding` folder contains a self-contained Python reference implementation (encoding.py) with test cases for multi-turn conversations, tool calling, thinking mode, numeric reasoning effort, mid-conversation system messages, and interleaved image content.

For production use, we additionally release deepseek-recipe, a set of Rust libraries with Python bindings that provides the same prompt format as a maintained, protocol-aware toolkit. It converts Messages, Chat Completions, and Responses API requests into the Conversation format, encodes them into DeepSeek V4 and V4.1 prompts or token IDs, and parses model output back into complete or streamed responses — covering thinking, tool calls, images, and generation settings. Model inference, tool execution, and HTTP transport are left to the caller.

Minimal Inference

Please refer to the `inference` folder for instructions on weight conversion and running inference locally.

Recommended sampling parameters:

| Parameter | Value | | --- | --- | | temperature | 1.0 | | top_p | 1.0 | | context_window | 1M tokens | | max_tokens | ≥ 256K |

License

This repository and the model weights are licensed under the MIT License.

Citation

code
@misc{deepseekai2026deepseekv41flash,
      title={DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression},
      author={DeepSeek-AI},
      year={2026},
}

Contact

If you have any questions, please raise an issue or contact us at [[email protected]](mailto:[email protected]).