DeepSeek-V4-Pro-0813
Description
DeepSeek-V4-Pro-0813 is DeepSeek-AI's official DeepSeek-V4-Pro release. It supersedes the preview model, adds a DSpark speculative-decoding module, and is designed for text generation, reasoning, coding, and agentic tool-use workflows.
This model is ready for commercial or non-commercial use.
Third-Party Community Consideration:
This model is not owned or developed by NVIDIA. This model has been developed and built to a third-party's requirements for this application and use case; see link to Non-NVIDIA DeepSeek AI DeepSeek-V4-Pro-0813 Model Card.
License and Terms of Use:
GOVERNING TERMS: This trial service is governed by the NVIDIA API Trial Terms of Service. Use of this model is governed by the NVIDIA Open Model Agreement. Additional Information: MIT.
Deployment Geography:
Global
Use Case:
Use Case: Text generation, reasoning, coding, and agentic tool-use workflows.
Release Date:
Build.NVIDIA.com: 08/24/2026 via link
Huggingface: 08/13/2026 via link
Reference(s):
References:
- DeepSeek-V4-Pro-0813 Model Page
- DeepSeek-V4 Technical Report
- DeepSeek Chat Interface
- vLLM Recipe for DeepSeek-V4-Pro
- SGLang Cookbook for DeepSeek-V4
Model Architecture:
Architecture Type: Transformer
Network Architecture: Sparse Mixture of Experts with hybrid Compressed Sparse Attention and Heavily Compressed Attention, Manifold-Constrained Hyper-Connections, and an attached DSpark speculative-decoding module
Total Parameters: 1.65T
Active Parameters: 49B
Vocabulary Size: 129,280
Input:
Input Types: Text
Input Formats: String
Input Parameters: One-Dimensional (1D)
Other Input Properties: Supports multi-turn messages encoded in OpenAI-compatible format and low, high, and max reasoning-effort levels.
Input Context Length (ISL): 1,000,000
Output:
Output Types: Text
Output Format: String
Output Parameters: One-Dimensional (1D)
Other Output Properties: Generates text completions and reasoning content.
Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA's hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.
Software Integration:
Runtime Engines:
- vLLM
- SGLang
Supported Hardware:
- NVIDIA Blackwell: B200, GB300
Preferred Operating Systems: Linux
The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment.
Model Version(s)
DeepSeek-V4-Pro-0813 v1.0
Training, Testing, and Evaluation Datasets:
Training Dataset
Data Modality: Text
Text Training Data Size: More than 10 Trillion Tokens
Training Data Collection: Undisclosed
Training Labeling: Undisclosed
Training Properties: The DeepSeek-V4 family was pretrained on more than 32 trillion diverse tokens and then post-trained for reasoning and agentic capabilities.
Testing Dataset
Testing Data Collection: Undisclosed
Testing Labeling: Undisclosed
Testing Properties: Undisclosed
Evaluation Dataset
Evaluation Data Collection: Hybrid: Automated, Manually-Collected
Evaluation Labeling: Hybrid: Automated, Manually-Labeled
Evaluation Properties: Evaluated on coding-agent, repository, cybersecurity, software-engineering, and tool-use benchmarks.
Evaluation Benchmark Score: DeepSeek-V4-Pro-0813 was evaluated across ten coding, repository, cybersecurity, software-engineering, and agentic tool-use benchmarks. Selected results include HLE with tools at 60.0, Terminal Bench 2.1 at 87.9, NL2Repo at 61.5, Cybergym at 83.3, DeepSWE at 62.7, and Toolathlon-Verified at 74.1.
View Detailed Benchmark Results
| Benchmark | DeepSeek-V4-Pro-0813 | DeepSeek-V4-Flash-0731 | DeepSeek-V4-Pro (Preview) | DeepSeek-V4-Flash (Preview) | GLM-5.2 | Kimi K3 | Opus-4.8 | Fable-5 (w/ fallback) |
|---|---|---|---|---|---|---|---|---|
| HLE (wo / w tools) | 42.7 / 60.0 | 37.8 / 51.5 | 37.7 / 48.2 | 34.8 / 45.1 | 40.5 / 54.7 | 43.5 / 56.0 | 49.8 / 57.9 | 53.3 / 63.0 |
| Terminal Bench 2.1 | 87.9 | 82.7 | 72.1 | 61.8 | 81.0 | 88.3 | 85.0 | 88.0 |
| NL2Repo | 61.5 | 54.2 | 38.5 | 39.4 | 48.9 | - | 69.7 | - |
| Cybergym | 83.3 | 76.7 | 52.7 | 38.7 | - | 80.0 | 78.3 | 83.1 |
| DeepSWE | 62.7 | 54.4 | 12.8 | 7.3 | 46.2 | 67.5 | 58.0 | 70.0 |
| Toolathlon-Verified | 74.1 | 70.3 | 55.9 | 49.7 | 59.9 | 76.5 | 76.2 | 77.9 |
| Agents' Last Exam | 25.7 | 25.2 | 16.5 | 15.8 | 23.8 | 27.6 | 25.7 | - |
| AutomationBench (Public) | 31.8 | 25.1 | 12.8 | 10.8 | 12.9 | 30.8 | 27.2 | 29.1 |
| DSBench-FullStack † | 71.1 | 68.7 | 41.8 | 37.0 | 61.8 | 73.7 | 71.6 | 77.2 |
| DSBench-Hard † | 67.2 | 59.6 | 31.1 | 25.8 | 54.5 | 63.0 | 71.7 | 68.3 |
Evaluation Methodology Notes:
- For the public code-agent benchmarks, DeepSeek-V4-Pro-0813 was evaluated with the minimal mode of DeepSeek Harness as the agent framework, using the max reasoning-effort level with
temperature = 1.0andtop_p = 0.95. - DSBench-FullStack is an internal full-stack development test set; DSBench-Hard is an internal test set of difficult coding-agent problems.
Inference
Acceleration Engine: vLLM, SGLang
Test Hardware: NVIDIA Blackwell (B200)
Ethical Considerations
NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. Developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.
Users are responsible for model inputs and outputs. Users are responsible for ensuring safe integration of this model, including implementing guardrails as well as other safety mechanisms, prior to deployment.
Please report model quality, risk, security vulnerabilities or NVIDIA AI Concerns here.
