Model Library
DeepSeekReleased August 21, 2025DeepSeek V3.1
DeepSeek · Released August 21, 2025
Intelligence score
21
At a glance
- Intelligence
- 21
- Speed
- —
- Latency
- —
- Input price
- $0.56per 1M tokens
- Output price
- $1.68per 1M tokens
- Context window
- 130Ktokens
- Parameters
- 685billions
Summary
Comparison summary
DeepSeek V3.1 (Non-reasoning) scores 21 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 17). DeepSeek V3.1 (Non-reasoning) costs $0.56 per 1M input tokens (better than average, median: $0.60) and $1.68 per 1M output tokens (better than average, median: $2.40), based on the median across providers serving the model.
Specifications
Technical specifications
| Model name | DeepSeek V3.1 |
|---|---|
| Provider | DeepSeek |
| Release date | August 21, 2025 |
| Intelligence score | 21 |
| Speed | — |
| Latency (TTFT) | — |
| Input price | $0.56 / 1M tokens |
| Output price | $1.68 / 1M tokens |
| Context window | 130,000 tokens |
| Parameters | 685B |
| Reasoning | No |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | MIT |
| API availability | — |
| Source | Model Intelligence legacy profile: deepseek-v3-1 |
Pricing
Token prices
Price per 1M tokens
DeepSeek V3.1, as listed in the catalog record
- Input$0.56
- Output$1.68
| Token type | Price per 1M |
|---|---|
| Input | $0.56 |
| Output | $1.68 |
Model Intelligence legacy profile: deepseek-v3-1
FAQ · 18
Frequently asked questions
DeepSeek V3.1 (Non-reasoning) was released on August 21, 2025.
DeepSeek V3.1 (Non-reasoning) was created by DeepSeek.
DeepSeek V3.1 (Non-reasoning) scores 21 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 17).
DeepSeek V3.1 (Non-reasoning) costs $0.56 per 1M input tokens (better than average, median: $0.60) and $1.68 per 1M output tokens (better than average, median: $2.40), based on the median across providers serving the model.
DeepSeek V3.1 (Non-reasoning) costs $0.56 per 1M input tokens and $1.68 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.67 per 1M tokens. Pricing may vary by provider.
No, DeepSeek V3.1 (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
DeepSeek V3.1 (Non-reasoning) supports text input.
DeepSeek V3.1 (Non-reasoning) supports text output.
No, DeepSeek V3.1 (Non-reasoning) does not support image input. It can only process text.
No, DeepSeek V3.1 (Non-reasoning) is not multimodal. It only supports text input.
DeepSeek V3.1 (Non-reasoning) has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, DeepSeek V3.1 (Non-reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
DeepSeek V3.1 (Non-reasoning) has 685 billion parameters (37 billion active).
DeepSeek V3.1 (Non-reasoning) is a Mixture of Experts (MoE) model with 685 billion total parameters, but only 37 billion active parameters are used during inference.
DeepSeek V3.1 (Non-reasoning) is released under the MIT license. This license allows commercial use.
DeepSeek V3.1 (Non-reasoning) achieves a score of 21 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, DeepSeek V3.1 (Non-reasoning) is available via API through 9 providers.
DeepSeek V3.1 (Non-reasoning) is available through 9 API providers.