DeepSeek-V3.1
DeepSeek · · 213 days after DeepSeek-R1
Verified
3 of 3 published values traced to a source3 primary sources, 3 archived
Why this status
Every value this record asserts — date (2025-08-21), context window (128K) and parameter count (671B) — was found in an archived primary source, checked 2026-08-10. Modalities recorded as text-to-text. The primary source states the model is text-only. Capabilities evidenced from the primary sources: reasoning: “5 Release 2024/09/05 Context Caching is Available 2024/08/02 New API Features 2024/07/25 API Reference API Guides Reasoning Model (deepseek-reasoner) Multi-round Conversa” | function_calling: “5 Release 2024/09/05 Context Caching is Available 2024/08/02 New API Features 2024/07/25 API Reference API Guides Reasoning Model (deepseek-reasoner) Multi-round Conversa” | tool_use: “DeepSeek-R1-0528 🛠️ Stronger agent skills: Post-training boosts tool use and multi-step agent tasks Try it now — toggle Think/Non-Think via the "DeepThink" but” | agentic: “DeepSeek-R1-0528 🛠️ Stronger agent skills: Post-training boosts tool use and multi-step agent tasks Try it now — toggle Think/Non-Think via the "DeepThink" but”.
What these states mean · How records are judged · Has this record changed?
A hybrid think-and-non-think update doubling context to 128K for the agent era.
Details
| Released | August 21, 2025stated in huggingface.coDeepSeek-V3.1 was released on August 21, 2025.
|
|---|---|
| Company | DeepSeek |
| Family | DeepSeek |
| Lineage | after DeepSeek-V3 |
| Era | Reasoning models |
| Type | Language model · LLM |
| Modalities | Text → Text |
| Capabilities | Agentic, Function calling, Reasoning, Tool use |
| Weights | Open weights |
| Licence | MIT |
| Context window | 128K tokensstated in api-docs.deepseek.com, huggingface.coDeepSeek-V3.1 supports a context window of 128,000 tokens.
|
| Parameters | 671Bstated in huggingface.coDeepSeek-V3.1 has 671,000,000,000 parameters.
|
Pricing
The price a lab's own page showed on the day that page was captured. That is what a snapshot can prove; it is not necessarily the launch price, and prices are often cut later. Citing the live pricing page instead would give a number with no date at all.
| Input | $0.27 per million tokens observed September 1, 2025 · archived snapshot (2025-09-01) |
|---|---|
| Output | $1.1 per million tokens observed September 1, 2025 · archived snapshot (2025-09-01) |
Current information, elsewhere
This record is a historical one: every figure above is traced to a dated primary source. For what is true today — live pricing, current rankings, availability — these are better maintained than anything here could be.
- Current price and providers OpenRouter ↗
- Current benchmark scores Artificial Analysis ↗
- Weights and model card Hugging Face ↗
Sources
Record status: verified · confidence 90/100 · 3 of 3 primary
Every value this record asserts — date (2025-08-21), context window (128K) and parameter count (671B) — was found in an archived primary source, checked 2026-08-10. Modalities recorded as text-to-text. The primary source states the model is text-only. Capabilities evidenced from the primary sources: reasoning: “5 Release 2024/09/05 Context Caching is Available 2024/08/02 New API Features 2024/07/25 API Reference API Guides Reasoning Model (deepseek-reasoner) Multi-round Conversa” | function_calling: “5 Release 2024/09/05 Context Caching is Available 2024/08/02 New API Features 2024/07/25 API Reference API Guides Reasoning Model (deepseek-reasoner) Multi-round Conversa” | tool_use: “DeepSeek-R1-0528 🛠️ Stronger agent skills: Post-training boosts tool use and multi-step agent tasks Try it now — toggle Think/Non-Think via the "DeepThink" but” | agentic: “DeepSeek-R1-0528 🛠️ Stronger agent skills: Post-training boosts tool use and multi-step agent tasks Try it now — toggle Think/Non-Think via the "DeepThink" but”.
- api-docs.deepseek.com Official documentation archived
- huggingface.co Model card archived
- api-docs.deepseek.com Official documentation archived
Primary means published by the organisation that made the model. Verified records require at least one.
DeepSeek family
- DeepSeek-V2 May 6, 2024
- DeepSeek-V2.5 September 5, 2024
- DeepSeek-V3 December 26, 2024
- DeepSeek-R1 January 20, 2025
- DeepSeek-V3.1 August 21, 2025
- DeepSeek V4 Flash April 24, 2026
- DeepSeek-V4.1-Flash September 10, 2026
What changed from DeepSeek-V3
Compared with the previous release in this series. A field appears only when both records state a value — where one does not, it is listed as uncomparable rather than dropped, because a gap in our research is not a finding about the model.
- LicenceDeepSeek License Agreement 1.0 → MIT
- Capabilities first evidencedagentic, function_calling, reasoning, tool_use
Compare DeepSeek-V3.1 with another model →
See DeepSeek-V3.1 on the interactive timeline →