DeepSeek model product
DeepSeek-V3.1
1 is a hybrid large language model with thinking and non-thinking modes, 671B total parameters, and 37B activated parameters.
Updated Sep 8, 2026. Default version: DeepSeek-V3.1
LLMBoard Score39.0DeepSeek-V3.1
Coverage80%16 benchmark families
Context window163.8KTokens
Official input priceN/AOfficial price unavailable
DeepSeek-V3.1 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
DeepSeek-V3.1 Benchmark Results
Benchmark scores for DeepSeek-V3.1.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkSimpleQA | Score93.40% | Rank03 | Participants47 | Percentile95.65% | EvidenceC | Evaluated |
| BenchmarkAider-Polyglot | Score68.40% | Rank08 | Participants22 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkBrowseComp-zh | Score49.20% | Rank10 | Participants13 | Percentile25.00% | EvidenceC | Evaluated |
| BenchmarkCodeForces | Score69.70% | Rank13 | Participants17 | Percentile25.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench | Score31.30% | Rank18 | Participants25 | Percentile29.17% | EvidenceC | Evaluated |
| BenchmarkMMLU-Redux | Score91.80% | Rank21 | Participants48 | Percentile57.45% | EvidenceC | Evaluated |
| BenchmarkHMMT 2025 | Score33.50% | Rank32 | Participants33 | Percentile3.13% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score83.70% | Rank33 | Participants138 | Percentile76.64% | EvidenceC | Evaluated |
| BenchmarkSWE-bench Multilingual | Score54.50% | Rank35 | Participants43 | Percentile19.05% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench | Score56.40% | Rank38 | Participants75 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkAIME 2024 | Score66.30% | Rank43 | Participants53 | Percentile19.23% | EvidenceC | Evaluated |
| BenchmarkBrowseComp | Score30.00% | Rank60 | Participants63 | Percentile4.84% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Verified | Score66.00% | Rank77 | Participants113 | Percentile32.14% | EvidenceC | Evaluated |
| BenchmarkHumanity's Last Exam | Score15.90% | Rank78 | Participants103 | Percentile24.51% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,420.30 rating | Rank79 | Participants210 | Percentile62.68% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,417.53 rating | Rank91 | Participants210 | Percentile56.94% | EvidenceA | Evaluated |
| BenchmarkAIME 2025 | Score49.80% | Rank105 | Participants119 | Percentile11.86% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score74.90% | Rank119 | Participants247 | Percentile52.03% | EvidenceC | Evaluated |
DeepSeek-V3.1 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categoryindustry medicine and healthcare | Rank29 | Rating / score1,466.91 | Votes227 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank35 | Rating / score1,459.24 | Votes257 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry legal and government | Rank48 | Rating / score1,462.76 | Votes257 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank51 | Rating / score1,434.67 | Votes284 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank54 | Rating / score1,452.24 | Votes572 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank57 | Rating / score1,469.97 | Votes1,150 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank57 | Rating / score1,429.27 | Votes208 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank60 | Rating / score1,448.22 | Votes286 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry writing and literature and language | Rank61 | Rating / score1,411.95 | Votes754 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry medicine and healthcare | Rank61 | Rating / score1,463.91 | Votes227 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank63 | Rating / score1,439.06 | Votes564 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank63 | Rating / score1,444.43 | Votes564 | ObservationsN/A | Result date |
| Arenatext style control | Categorylonger query | Rank64 | Rating / score1,448.00 | Votes672 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank65 | Rating / score1,402.16 | Votes450 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank69 | Rating / score1,403.92 | Votes450 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry writing and literature and language | Rank69 | Rating / score1,411.84 | Votes754 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank69 | Rating / score1,380.55 | Votes329 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank70 | Rating / score1,428.47 | Votes409 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry life and physical and social science | Rank70 | Rating / score1,452.93 | Votes572 | ObservationsN/A | Result date |
| Arenatext style control | Categoryrussian | Rank71 | Rating / score1,429.20 | Votes284 | ObservationsN/A | Result date |
| Arenatext | Categorylonger query | Rank72 | Rating / score1,424.97 | Votes672 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank72 | Rating / score1,471.73 | Votes1,150 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank72 | Rating / score1,419.64 | Votes208 | ObservationsN/A | Result date |
| Arenatext style control | Categoryspanish | Rank72 | Rating / score1,421.78 | Votes409 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank73 | Rating / score1,410.78 | Votes397 | ObservationsN/A | Result date |
| Arenatext style control | Categoryinstruction following | Rank74 | Rating / score1,421.36 | Votes853 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank75 | Rating / score1,417.45 | Votes689 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank75 | Rating / score1,374.20 | Votes329 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank75 | Rating / score1,447.12 | Votes286 | ObservationsN/A | Result date |
| Arenatext | Categoryinstruction following | Rank76 | Rating / score1,407.32 | Votes853 | ObservationsN/A | Result date |
DeepSeek-V3.1 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $0.19 input, $0.71 output per 1M via TokenGo
- Tracked offerings
- 20
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderTokenGo | Provider model IDdeepseek/deepseek-v3.1 | Regionglobal | Input / 1M$0.19 | Output / 1M$0.71 | Context131.1K | Updated |
| ProviderNanoGPT | Provider model IDdeepseek-ai/DeepSeek-V3.1 | Regionglobal | Input / 1M$0.20 | Output / 1M$0.70 | Context128K | Updated |
| Providersubmodel | Provider model IDdeepseek-ai/DeepSeek-V3.1 | Regionglobal | Input / 1M$0.20 | Output / 1M$0.80 | Context75K | Updated |
| ProviderDeep Infra | Provider model IDdeepseek-ai/DeepSeek-V3.1 | Regionglobal | Input / 1M$0.25 | Output / 1M$0.95 | Context163.8K | Updated |
| ProviderOpenRouter | Provider model IDdeepseek/deepseek-chat-v3.1 | Regionglobal | Input / 1M$0.25 | Output / 1M$0.95 | Context163.8K | Updated |
| ProviderVercel AI Gateway | Provider model IDdeepseek/deepseek-v3.1 | Regionglobal | Input / 1M$0.25 | Output / 1M$0.95 | Context163.8K | Updated |
| ProviderMeganova | Provider model IDdeepseek-ai/DeepSeek-V3.1 | Regionglobal | Input / 1M$0.27 | Output / 1M$1 | Context164K | Updated |
| ProviderHugging Face | Provider model IDdeepseek-ai/DeepSeek-V3.1 | Regionglobal | Input / 1M$0.27 | Output / 1M$1 | Context131.1K | Updated |
| ProviderJiekou.AI | Provider model IDdeepseek/deepseek-v3.1 | Regionglobal | Input / 1M$0.27 | Output / 1M$1 | Context163.8K | Updated |
| ProviderNovitaAI | Provider model IDdeepseek/deepseek-v3.1 | Regionglobal | Input / 1M$0.27 | Output / 1M$1 | Context131.1K | Updated |
| ProviderMerge Gateway | Provider model IDdeepseek/deepseek-v3.1 | Regionglobal | Input / 1M$0.50 | Output / 1M$1.5 | Context164K | Updated |
| ProviderBaseten | Provider model IDdeepseek-ai/DeepSeek-V3.1 | Regionglobal | Input / 1M$0.50 | Output / 1M$1.5 | Context164K | Updated |
| ProviderWeights & Biases | Provider model IDdeepseek-ai/DeepSeek-V3.1 | Regionglobal | Input / 1M$0.55 | Output / 1M$1.65 | Context161K | Updated |
| ProviderAbacus | Provider model IDdeepseek/deepseek-v3.1 | Regionglobal | Input / 1M$0.55 | Output / 1M$1.66 | Context128K | Updated |
| ProviderPioneer | Provider model IDdeepseek-ai/DeepSeek-V3.1 | Regionglobal | Input / 1M$0.56 | Output / 1M$1.68 | Context163.8K | Updated |
| ProviderAlibaba (China) | Provider model IDdeepseek-v3-1 | Regionglobal | Input / 1M$0.574 | Output / 1M$1.72 | Context131.1K | Updated |
| ProviderAmazon Bedrock | Provider model IDdeepseek.v3-v1:0 | Regionglobal | Input / 1M$0.58 | Output / 1M$1.68 | Context163.8K | Updated |
| ProviderVertex | Provider model IDdeepseek-ai/deepseek-v3.1-maas | Regionglobal | Input / 1M$0.60 | Output / 1M$1.7 | Context163.8K | Updated |
| ProviderTogether AI | Provider model IDdeepseek-ai/DeepSeek-V3-1 | Regionglobal | Input / 1M$0.60 | Output / 1M$1.7 | Context131.1K | Updated |
DeepSeek-V3.1 Runtime Performance
Provider-specific output speed and catalog latency for DeepSeek-V3.1. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
DeepSeek-V3.1 Specifications
Technical details for the model's default version.
- Version
- DeepSeek-V3.1
- Released
- Jan 10, 2025
- Knowledge cutoff
- Unknown
- Parameters
- 671B
- Context window
- 163.8K
- Max output
- 163.8K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- MIT
DeepSeek-V3.1 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionDeepSeek-V3.1 | Released | LLMBoard39.05 | Parameters671B | Context163.8K | Max output163.8K | Open weightsYes | LicenseMIT |
Models similar to DeepSeek-V3.1
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
DE
DeepSeek-R1
DeepSeek
DE
DeepSeek-V3
DeepSeek
DE
DeepSeek-V3.2
DeepSeek
DE
DeepSeek-R1-Distill-Qwen
DeepSeek
DE
DeepSeek-V2.5
DeepSeek
DE
DeepSeek-R1-Distill-Llama
DeepSeek
What is DeepSeek-V3.1?
Key information about DeepSeek-V3.1 and its available data.
1-Base that supports thinking and non-thinking modes through different chat templates. It uses a two-phase long context extension, supports up to 128K context, and uses the UE8M0 FP8 scale data format for model weights and activations.
Data as of 2026-09-08.
FAQ
Common questions about DeepSeek-V3.1.
When was DeepSeek-V3.1 released?
DeepSeek-V3.1's default version was released on Jan 10, 2025.
How much does DeepSeek-V3.1 cost?
No official standard PAYG price is currently available for DeepSeek-V3.1. The lowest tracked third-party offer starts at $0.19 input and $0.71 output via TokenGo.
Who created DeepSeek-V3.1?
DeepSeek-V3.1 was created by DeepSeek.
What is the context window for DeepSeek-V3.1?
The default version has a 163.8K token context window.
Is DeepSeek-V3.1 open weight?
Yes. The default version is marked as open weight under MIT.
How many API providers offer DeepSeek-V3.1?
20 provider offerings are linked to the default version.