> ## Content Index
> Fetch the complete content index at: https://www.theleftshift.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# DeepSeek V3.1 Launches Quietly, But Signals Big Strides for Open-Source AI
- URL: https://www.theleftshift.com/deepseek-v3-1-launches-quietly-but-signals-big-strides-for-open-source-ai/
- Published: 2025-08-21T08:32:20.000Z
- Updated: 2025-08-21T08:32:20.000Z
- Description: The standout feature is its 128,000-token context length.
- Author: The Left Shift Bureau
- Tags: AI Models, Open Source

DeepSeek has quietly [released](https://huggingface.co/deepseek-ai/DeepSeek-V3.1?ref=theleftshift.com) **V3.1**, an upgraded version of its large language model, on **August 19, 2025**, via its official WeChat group. Despite the low-key announcement, the update is making waves in the AI community for pushing the boundaries of open-source performance.

The standout feature is its **128,000-token context length**, allowing V3.1 to handle lengthy technical documents, conversations, and retrieval tasks more effectively. 

Only last week, Anthropic rolled out long-context support for Claude Sonnet 4, now capable of [processing up to 1 million tokens](https://www.theleftshift.com/claude-sonnet-4-gets-1-mn-token-context-support-enabling-entire-codebase-and-multi-doc-processing/)—a 5x increase—on the Anthropic API. This breakthrough allows developers to feed in full codebases (over 75,000 lines of code) or analyse dozens of documents in a single prompt.

With **685 billion parameters**, DeepSeek's newer model builds on the previous V3’s architecture, but its **Mixture-of-Experts** design ensures that only 37 billion parameters are active per token—keeping costs low and efficiency high.

V3.1 shows marked improvements in **coding, logic, and math**. In community tests, it outperformed peers in Python and Bash tasks, while also achieving high accuracy in problem-solving and mathematical reasoning. It continues to surpass rival models like Qwen2.5 72B in benchmarks such as AIME and MATH-500.

Released under the **MIT License**, DeepSeek V3.1 is available on **Hugging Face**, with full support for Safetensors. Training costs came in at just **$5.6 million**, a fraction of the budget for closed models. Developers already see it as a turning point—an open model that’s beginning to **rival GPT-4o and Claude 3.5 Sonnet**.

Reportedly, he update also [removed R1 references](https://www.scmp.com/tech/big-tech/article/3322481/deepseeks-v31-update-and-missing-r1-label-spark-speculation-over-fate-r2-ai-model?ref=theleftshift.com), sparking speculation about a shift from reasoning research. The release wasn’t announced on the company’s public platforms, including its official X account.

Earluer this year, the startup released a [minor update to its flagship R1 reasoning model](https://www.theleftshift.com/deepseek-releases-an-updated-r1-model-on-hugging-face/) on the developer platform Hugging Face.