LLM
DeepSeek-V3.1
DeepSeek hybrid model with thinking and non-thinking modes and stronger tool use and agent skills, post-trained on a long-context extension of the V3 base.
Levelmodel
ClassLLM
Parameters671B
Context128,000
Released21 Aug 2025
Opennessopen weights permissive
Runs onnot in the notes
Statusactive
Benchmark Results
vendor-reported
AIME 2025 2025math
Thinking mode, Pass@1vendor-reported
88.4
050100
Score in %. Higher is better.
DeepSeek-V3.1 model card
BrowseCompagents and tool use
Thinking mode, internal search agent frameworkvendor-reported
30
050100
Score in %. Higher is better.
DeepSeek-V3.1 model card
GPQA Diamondknowledge and reasoning
Thinking mode, Pass@1vendor-reported
80.1
050100
Score in %. Higher is better.
DeepSeek-V3.1 model card
LiveCodeBenchcoding
Thinking mode, 2408 to 2505, Pass@1vendor-reported
74.8
050100
Score in %. Higher is better.
DeepSeek-V3.1 model card
MMLU-Proknowledge and reasoning
Thinking mode, EMvendor-reported
84.8
050100
Score in %. Higher is better.
DeepSeek-V3.1 model card
SWE-bench Multilingualcoding
Non-thinking, agent modevendor-reported
54.5
050100
Score in %. Higher is better.
DeepSeek-V3.1 model card
SWE-bench Verifiedcoding
Non-thinking, agent modevendor-reported
66
050100
Score in %. Higher is better.
DeepSeek-V3.1 model card
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).
Lineage
successorcontinued pretrain
DeepSeek-V3.1-Terminus is a child (successor)high confidencepress claimDeepSeek v3 is the parent (continued pretrain)high confidencemodel cardWhere this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 3 models. Confidence and evidence are the notes' own.
R&D Topics
Identity
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.
Architecture and Size
- Architecture
- MoE Hybrid thinking and non-thinking modes in one model; trained with UE8M0 FP8 scale format
- Total parameters
- 671B sources
- Active parameters
- 37B
- Checkpoint parameters
- –
- Context length
- 128,000 tokens
- Input
- text
- Output
- text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).
Openness and Access
- Openness
- open weights permissive
- Licence
- MIT License sources
- Access routes
- API, consumer app, open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.
Deployment
- Runs on
- not in the notes
- Footprint
- not in the notes
- Runtimes
- not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.
Indonesia
- Indonesian support
- not stated
- Regional languages
- not in the notes
- Jakarta hosting
- unknown
- Languages
- not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.
Industry Use Cases
No industry use case links this model.
Links
DeveloperDeepSeekstartup, China
DeepSeekstartup, ChinaModelDeepSeek-V3.1DeepSeek, LLM
DeepSeek-V3.1DeepSeek, LLMR&D TopicsAgents and Tool UseEfficiency Architectures (MoE, Sparse Attention, Long Context)Open-Weight Frontier Releases as Research SubstrateReasoning via Large-Scale RL
Use CasesProcess Optimisation and Advanced Process ControlIndustrialClinical LLM (Open Weights)HealthcareTelecom Domain LLMs and BenchmarksTelecomChinese Medical LLMsHealthcareUse cases of models in the same R&D topic; none is recorded for this model itself.
Who Runs It
PT Amman Mineral Nusa Tenggara (AMMAN)In production, 8 Jul 2026MIND IDResearch, 14 Oct 2025
Conch CementIn production, 23 Jun 2025
PT Kalbe Morinaga IndonesiaIn production, Oct 2024+2 moreDeployments of these use cases; none names this model.
PT Amman Mineral Nusa Tenggara (AMMAN)In production, 8 Jul 2026MIND IDResearch, 14 Oct 2025Conch CementIn production, 23 Jun 2025PT Kalbe Morinaga IndonesiaIn production, Oct 2024+2 moreDeployments of these use cases; none names this model.Organisations 1
R&D topics 4
Open-Weight Frontier Releases as Research SubstrateUsing released weights as the base for downstream research.used byEfficiency Architectures (MoE, Sparse Attention, Long Context)Reducing training and inference cost per capability.produced byReasoning via Large-Scale RLReinforcement learning on outcomes, often verifiable math or code, to produce long reason…produced byAgents and Tool UseSystems that plan, call tools and act over multi-step tasks.produced by
Sources
- huggingface.co/deepseek-ai/DeepSeek-V3.1 https://huggingface.co/deepseek-ai/DeepSeek-V3.1
- api-docs.deepseek.com/news/news250821 https://api-docs.deepseek.com/news/news250821
- openrouter.ai/deepseek/deepseek-chat-v3.1 https://openrouter.ai/deepseek/deepseek-chat-v3.1
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "DeepSeek-V3.1", https://aurion.semai.id/models/deepseek-deepseek-v3-1/, records of October 2026.