# DeepSeek-V3.1

> One open-weights model, two modes — fast answers or deep reasoning

- Maker: DeepSeek
- Line: DeepSeek V3
- Canonical page: https://ai-tldr.dev/models/deepseek-v3-1/
- Updated: 2026-06-18

## In short

DeepSeek-V3.1 is the August 2025 update to DeepSeek's V3 line, released on 21 August 2025 under an MIT license with open weights on Hugging Face. It has a 128K-token context window and costs $0.21 per million input tokens and $0.79 per million output tokens. You can use it through DeepSeek, OpenRouter and Hugging Face.

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Context window | 128K | [1] |
| Input price | $0.21 / 1M tokens | [2] |
| Output price | $0.79 / 1M tokens | [2] |
| Licence | MIT · open weights | [1] |
| Released | 21 August 2025 | [3] |
| Parameters | 671B total / 37B active (MoE) | [1] |

## Sources

1. [DeepSeek-V3.1 model card (Hugging Face)](https://huggingface.co/deepseek-ai/DeepSeek-V3.1)
2. [DeepSeek V3.1 on OpenRouter (pricing & API)](https://openrouter.ai/deepseek/deepseek-chat-v3.1)
3. [DeepSeek-V3.1 release notes (official)](https://api-docs.deepseek.com/news/news250821)

## Overview

DeepSeek-V3.1 is the August 2025 update to DeepSeek's V3 line, released on 21 August 2025 under an MIT license with open weights on Hugging Face. It keeps the same 671B-parameter Mixture-of-Experts backbone as DeepSeek-V3 (about 37B parameters active per token) and adds a hybrid design: a single model that can answer directly or switch into a chain-of-thought "thinking" mode. On DeepSeek's own API the two behaviours are exposed as deepseek-chat (non-thinking) and deepseek-reasoner (thinking), both sharing a 128K-token context window.

The headline change in DeepSeek-V3.1 is that one model now covers both quick replies and deep reasoning, where the previous generation split this across DeepSeek-V3-0324 and the R1 reasoning model. In thinking mode it reaches reasoning quality close to DeepSeek-R1-0528 while producing answers faster, and DeepSeek extended the long-context training (a 32K phase grown to 630B tokens and a 128K phase to 209B tokens) so the model holds up better over long inputs.

DeepSeek-V3.1 also targets agent workloads. DeepSeek post-trained it for stronger tool calling, multi-step search agents, and code agents, and reports gains on coding and terminal benchmarks over the prior V3 and R1 releases. The model is text-only and ships with FP8 (UE8M0) weights for efficient inference, and the open MIT weights mean it can be self-hosted as well as called through DeepSeek's API and third-party providers.

## Benchmarks

- MMLU-Redux (Thinking): 93.7% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- MMLU-Pro (Thinking): 84.8% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- GPQA-Diamond (Thinking): 80.1% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- AIME 2025 (Thinking): 88.4% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- AIME 2024 (Thinking): 93.1% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- HMMT 2025 (Thinking): 84.2% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- LiveCodeBench (Thinking): 74.8% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- Aider-Polyglot (Thinking): 76.3% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- SWE-bench Verified (Agent): 66% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- SWE-bench Multilingual (Agent): 54.5% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- Terminal-bench: 31.3% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- BrowseComp (Thinking): 30% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))
- Humanity's Last Exam (Thinking): 15.9% ([source](https://huggingface.co/deepseek-ai/DeepSeek-V3.1))

## Pricing

- Input: $0.21 / 1M tokens per 1M tokens
- Output: $0.79 / 1M tokens per 1M tokens

Pricing for deepseek-chat-v3.1 as listed by OpenRouter. DeepSeek's first-party API has since migrated this model line to newer model names, so check the provider for live rates.

Source: https://openrouter.ai/deepseek/deepseek-chat-v3.1

## Strengths

- Hybrid think/non-think in one model — pick fast answers or deep reasoning without swapping models
- Open weights under a permissive MIT license, so it can be self-hosted and fine-tuned
- Strong reasoning: ~88 on AIME 2025 and 80 on GPQA-Diamond in thinking mode
- Tuned for agents — better tool calling, code agents, and multi-step search than prior V3/R1 releases
- Large 128K context window with extended long-context training
- Very low API pricing relative to closed frontier models

## Best for

- Coding agents and software-engineering tasks (SWE-bench-style fix/patch workflows)
- Hard math and reasoning problems where the thinking mode pays off
- Multi-step research and tool-using/search agents
- Cost-sensitive high-volume chat and generation in non-thinking mode
- Self-hosted deployments that need open weights and a permissive license
- Long-document analysis within the 128K context window

## How to access

- [DeepSeek](https://api-docs.deepseek.com/news/news250821) — model id `deepseek-chat / deepseek-reasoner`
- [OpenRouter](https://openrouter.ai/deepseek/deepseek-chat-v3.1) — model id `deepseek/deepseek-chat-v3.1`
- [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V3.1) — model id `deepseek-ai/DeepSeek-V3.1`

## DeepSeek V3 versions

- [DeepSeek-V3.2](https://ai-tldr.dev/models/deepseek-v3-2/) (1 December 2025)
- [DeepSeek-V3.2-Speciale](https://ai-tldr.dev/models/deepseek-v3-2-speciale/) (1 December 2025)
- [DeepSeek-V3.2-Exp](https://ai-tldr.dev/models/deepseek-v3-2-exp/) (29 September 2025)
- [DeepSeek-V3.1-Terminus](https://ai-tldr.dev/models/deepseek-v3-1-terminus/) (22 September 2025)
- DeepSeek-V3.1 (21 August 2025) — this page
- [DeepSeek-V3-0324](https://ai-tldr.dev/models/deepseek-v3-0324/) (24 March 2025)
- [DeepSeek-V3](https://ai-tldr.dev/models/deepseek-v3/) (26 December 2024)
- [DeepSeek-V2.5](https://ai-tldr.dev/models/deepseek-v2-5/) (5 September 2024)
- [DeepSeek-V2](https://ai-tldr.dev/models/deepseek-v2/) (May 2024)

## FAQ

### Is DeepSeek-V3.1 open source?

The weights are openly released on Hugging Face under an MIT license, so you can self-host, fine-tune, and use the model commercially. It is open-weights; DeepSeek does not publish the full training dataset.

### What is hybrid thinking mode in DeepSeek-V3.1?

DeepSeek-V3.1 is a single model with two modes. A non-thinking mode (deepseek-chat) answers directly for speed, and a thinking mode (deepseek-reasoner) produces chain-of-thought reasoning for harder problems. Both share the same 128K-token context window.

### How big is DeepSeek-V3.1 and how much context does it support?

It is a 671B-parameter Mixture-of-Experts model that activates about 37B parameters per token, with a 128K-token context window. It is text-only and uses Multi-head Latent Attention with FP8 (UE8M0) weights.

### When was DeepSeek-V3.1 released?

DeepSeek-V3.1 was released on 21 August 2025 as the hybrid-reasoning update to the DeepSeek-V3 line.

## Links

- [DeepSeek-V3.1 release notes (official)](https://api-docs.deepseek.com/news/news250821)
- [DeepSeek-V3.1 model card (Hugging Face)](https://huggingface.co/deepseek-ai/DeepSeek-V3.1)
- [DeepSeek API documentation](https://api-docs.deepseek.com/quick_start/pricing)
- [DeepSeek V3.1 on OpenRouter (pricing & API)](https://openrouter.ai/deepseek/deepseek-chat-v3.1)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/deepseek-v3-1/