The NVIDIA Qwen3-235B-A22B-Thinking-2507-FP4 Eagle model is the Eagle head of the Alibaba’s Qwen3-235B-A22B-Thinking-2507 model in FP4, which is an auto-regressive language model that uses a mixture-of-experts (MoE) architecture with 32 billion activated parameters and 1 trillion total parameters. For more information, please check here. The NVIDIA Qwen3-235B-A22B-Thinking-2507-FP4 Eagle3 model incorporates Eagle speculative decoding with TensorRT Model Optimizer.
This model is ready for commercial/non-commercial use.
Developers designing AI Agent systems, chatbots, RAG systems, and other AI-powered applications. Also suitable for typical instruction-following tasks.
Architecture Type: Transformers Network Architecture: Llama3 Model Parameters: 1B This model was developed based on Qwen3-235B-A22B-Thinking-2507-FP4
Input:
Input Type(s): Text Input Format(s): String Input Parameters: One-Dimensional (1D): Sequences Other Properties Related to Input: Max Context Length: 262144
Output:
Output Type(s): Text Output Format: String Output Parameters: One-Dimensional (1D): Sequences Other Properties Related to Output: Max Context Length: 262144
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.
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):
v1
Training and Evaluation Datasets:
** The total size (in number of data points): 6.34M
** Total number of datasets: 1
** Dataset partition: Training 100%
Training Dataset:
Link: From the Nemotron-Post-Training-Dataset-v2, only prompts were used for data synthesis; the original responses from GPT were not used. The synthesized data was then used to train the Eagle modules.
Properties: 80 multi-turn dialogue sequences, each annotated with expert preference votes.
Inference:
Acceleraton Engine: TensorRT-LLM 1.2.0rc6 Test Hardware: B200
Eagle Speculative Decoding
Synthesized data was obtained from Alibaba's Qwen3-235B-A22B-Thinking-2507 model, which is then used to finetune the Eagle modules. This model is ready for inference with TensorRT-LLM in Eagle speculative decoding mode. Eagle modules are used to predict candidate tokens beyond the next token. In the generation step, each forward Eagle module generates a distribution of tokens beyond the previous. Then, a tree-based attention mechanism samples some candidate sequences for the original model to validate. The longest accepted candidate sequence is selected so that more than 1 token is returned in the generation step. The number of tokens generated in each step is called acceptance rate.
Usage
To serve the checkpoint with TensorRT-LLM, follow the sample commands below with the TensorRT-LLM GitHub repo:
The Eagle acceptance rate benchmark results (MT-Bench) with draft length 3 are presented in the table below:
Category
MT Bench Acceptance Rate
writing
2.13
roleplay
2.05
reasoning
2.49
math
3.15
coding
2.68
extraction
2.72
stem
2.27
humanities
2.06
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. When downloaded or used in accordance with our terms of service, 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.
For more detailed information on ethical considerations for this model, please see the Model Card++ Explainability, Bias, Safety & Security, and Privacy Subcards below.
Please report model quality, risk, security vulnerabilities or NVIDIA AI Concerns here.
SUBCARDS:
Explainability
Field:
Response:
Intended Task/Domain:
Text generation, reasoning, summarization, and question answering.
Model Type:
Text and Image-to-text transformer
Intended Users:
This model is intended for developers, researchers, and customers building/utilizing LLMs, while balancing accuracy and efficiency.
Output:
Text String(s)
Describe how the model works:
Generates text by predicting the next word or token based on the context provided in the input sequence using multiple self-attention layers.
Technical Limitations & Mitigation:
The model was trained on data that contains toxic language and societal biases originally crawled from the internet. Therefore, the model may amplify those biases and return toxic responses especially when prompted with toxic prompts. Therefore, before deploying any applications of this model, developers should perform safety testing and tuning tailored to their specific applications of the model.
Verified to have met prescribed quality standards?
Yes
Performance Metrics:
Accuracy, Throughput, and user-side throughput
Potential Known Risk
The model may generate answers that may be inaccurate, omit key information, or include irrelevant or redundant text producing socially unacceptable or undesirable text, even if the prompt itself does not include anything explicitly offensive.
The Principle of least privilege (PoLP) is applied limiting access for dataset generation. Restrictions enforce dataset access during training, and dataset license constraints adhered to. Model checkpoints are made available on Hugging Face, and may become available on cloud providers' model catalog.
Privacy
Field:
Response:
Generatable or reverse engineerable personal data?
No
Was consent obtained for any personal data used?
Not Applicable
Personal data used to create this model?
None Known
How often is dataset reviewed?
Before Release
Was data from user interactions with the AI model (e.g. user input and prompts) used to train the model?
No
Is there provenance for all datasets used in training?
Yes
Does data labeling (annotation, metadata) comply with privacy laws?
Yes
Is data compliant with data subject requests for data correction or removal, if such a request was made?
