--- title: "cosmos-predict1-7b" publisher: "nvidia" type: "endpoint" updated: "2025-06-11T23:39:18.176Z" description: "Generalist model to generate future world state as videos from text and image prompts to create synthetic training data for robots and autonomous vehicles." canonical: "https://build.nvidia.com/nvidia/cosmos-predict1-7b" --- # **Cosmos-Predict1**: A Suite of Diffusion-Based World Foundation Models [**Cosmos**](https://huggingface.co/collections/nvidia/cosmos-6751e884dc10e013a0a0d8e6) | [**Code**](https://github.com/NVIDIA/Cosmos) | [**Paper**](https://research.nvidia.com/publication/2025-01_cosmos-world-foundation-model-platform-physical-ai) # Model Overview ## Description: **Cosmos World Foundation Models**: A family of highly performant pre-trained world foundation models purpose-built for accelerating synthetic data generation. Cosmos world foundation models include: 1. Cosmos Predict: Generalist model for predictive video generation from text, image, or video prompts that produces future frame sequences based on input context. Trained on 20M hours of physical AI data, the model serves as a strong foundation for post-training into specialized models for autonomous systems. 2. Cosmos Transfer: Multicontrol model to generate videos conditioned on ground-truth simulations or structured video inputs for physical accuracy, enabling amplification to diverse environments and lighting conditions. 3. Cosmos Reason: Multimodal reasoning model for planning text response based on spatial and temporal understanding from input video. **Model Developer**: NVIDIA ## Model Versions The Cosmos Predict 1 release, includes the following models: * [Cosmos-Predict1-7B-Text2World](https://huggingface.co/nvidia/Cosmos-Predict1-7B-Text2World) * Given a text description, predict an output video of 121 frames. * [Cosmos-Predict1-7B-Video2World](https://huggingface.co/nvidia/Cosmos-Predict1-7B-Video2World) * Given a text description and an image as the first frame, predict the next 120 frames. ### License: This model is released under the [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/). For a custom license, please contact [cosmos-license@nvidia.com](mailto:cosmos-license@nvidia.com). Under the NVIDIA Open Model License, NVIDIA confirms: * Models are commercially usable. * You are free to create and distribute Derivative Models. * NVIDIA does not claim ownership of any outputs generated using the Models or Derivative Models. **Important Note**: If you bypass, disable, reduce the efficacy of, or circumvent any technical limitation, safety guardrail or associated safety guardrail hyperparameter, encryption, security, digital rights management, or authentication mechanism contained in the Model, your rights under [NVIDIA Open Model License Agreement](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/) will automatically terminate. ## Model Architecture: **Cosmos-Predict1-7B-Text2World** and **Cosmos-Predict1-7B-Video2World** are diffusion transformer models designed for video denoising in the latent space. The network is composed of interleaved self-attention, cross-attention, and feedforward layers as its building blocks. The cross-attention layers allow the model to condition on input text throughout the denoising process. Before each layer, adaptive layer normalization is applied to embed the time information for denoising. When an image or video is provided as input, their latent frames are concatenated with the generated frames along the temporal dimension. Augment noise is added to conditional latent frames to bridge the training and inference gap. ## Cosmos-Predict1-7B-Text2World Input/Output Specifications * **Input** * **Input Type(s)**: Text * **Input Format(s)**: String * **Input Parameters**: One-dimensional (1D) * **Other Properties Related to Input**: * The input string should contain fewer than 300 words and should provide descriptive content for world generation, such as a scene description, key objects or characters, background, and any specific actions or motions to be depicted within the 5-second duration. * **Output** * **Output Type(s)**: Video * **Output Format(s)**: mp4 * **Output Parameters**: Three-dimensional (3D) * **Other Properties Related to Output**: The generated video will be a 5-second clip with a resolution of 1280x704 pixels at 24 frames per second (fps). The content of the video will visualize the input text description as a short animated scene, capturing the main elements mentioned in the input within the time constraints. ## Cosmos-Predict1-7B-Video2World Input/Output Specifications * **Input** * **Input Type(s)**: Text+Image, Text+Video * **Input Format(s)**: * Text: String * Image: jpg, png, jpeg, webp * Video: mp4 * **Input Parameters**: * Text: One-dimensional (1D) * Image: Two-dimensional (2D) * Video: Three-dimensional (3D) * **Other Properties Related to Input**: * The input string should contain fewer than 300 words and should provide descriptive content for world generation, such as a scene description, key objects or characters, background, and any specific actions or motions to be depicted within the 5-second duration. * The input image should be of 1280x704 resolution. * The input video should be of 1280x704 resolution and 9 input frames. * **Output** * **Output Type(s)**: Video * **Output Format(s)**: mp4 * **Output Parameters**: Three-dimensional (3D) * **Other Properties Related to Output**: The generated video will be a 5-second clip with a resolution of 1280x704 pixels at 24 frames per second (fps). The content of the video will use the provided image as the first frame and visualize the input text description as a short animated scene, capturing the main elements mentioned in the input within the time constraints. ## Software Integration **Runtime Engine(s):** * [Cosmos](https://github.com/NVIDIA/Cosmos) **Supported Hardware Microarchitecture Compatibility:** * NVIDIA Blackwell * NVIDIA Hopper * NVIDIA Ampere **Note**: We have only tested inference with BF16 precision. **Operating System(s):** * Linux (We have not tested it on other operating systems.) # Usage * See [Cosmos](https://github.com/NVIDIA/Cosmos) for details. # Evaluation Please see our [technical paper](https://research.nvidia.com/publication/2025-01_cosmos-world-foundation-model-platform-physical-ai) for detailed evaluations. ## Inference Time and GPU Memory Usage The numbers provided below may vary depending on system specs and are for reference only. We report the maximum observed GPU memory usage during end-to-end inference. Additionally, we offer a series of model offloading strategies to help users manage GPU memory usage effectively. For GPUs with limited memory (e.g., RTX 3090/4090 with 24 GB memory), we recommend fully offloading all models. For higher-end GPUs, users can select the most suitable offloading strategy considering the numbers provided below. ### Cosmos-Predict1-7B-Text2World | Offloading Strategy | 7B Text2World | 14B Text2World | |-------------|---------|---------| | Offload prompt upsampler | 74.0 GB | > 80.0 GB | | Offload prompt upsampler and guardrails | 57.1 GB | 70.5 GB | | Offload prompt upsampler, guardrails and T5 encoder | 38.5 GB | 51.9 GB | | Offload prompt upsampler, guardrails, T5 encoder and tokenizer | 38.3 GB | 51.7 GB | | Offload prompt upsampler, guardrails, T5 encoder, tokenizer and diffusion model | 24.4 GB | 39.0 GB | The table below presents the end-to-end inference runtime on a single H100 GPU, excluding model initialization time. | 7B Text2World (offload prompt upsampler) | 14B Text2World (offload prompt upsampler, guardrails) | |---------|---------| | ~380 seconds | ~590 seconds | ### Cosmos-Predict1-7B-Video2World | Offloading Strategy | 7B Video2World | 14B Video2World | |----------------------------------------------------------------------------------|---------|---------| | Offload prompt upsampler | 76.5 GB | > 80.0 GB | | Offload prompt upsampler and guardrails | 59.9 GB | 73.3 GB | | Offload prompt upsampler, guardrails and T5 encoder | 41.3 GB | 54.8 GB | | Offload prompt upsampler, guardrails, T5 encoder and tokenizer | 41.1 GB | 54.5 GB | | Offload prompt upsampler, guardrails, T5 encoder, tokenizer and diffusion model | 27.3 GB | 39.0 GB | The following table shows the end-to-end inference runtime on a single H100 GPU, excluding model initialization time: | 7B Video2World (offload prompt upsampler) | 14B Video2World (offload prompt upsampler, guardrails) | |---------|---------| | ~383 seconds | ~593 seconds | ## 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 subcards of Explainability, Bias, Safety & Security, and Privacy below. Please report security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/). ## Bias Field | Response :---------------------------------------------------------------------------------------------------|:--------------- Participation considerations from adversely impacted groups [protected classes](https://www.senate.ca.gov/content/protected-classes) in model design and testing: | None Measures taken to mitigate against unwanted bias: | None ## Explainability Field | Response :------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------- Intended Application and Domain: | World generation Model Type: | Transformer Intended Users: | Physical AI developers Output: | Videos Describe how the model works: | Generates videos based on video inputs. Technical Limitations: | The model may not follow the video input accurately. Verified to have met prescribed NVIDIA quality standards: | Yes Performance Metrics: | Quantitative and qualitative evaluation Potential Known Risks: | The model's output can generate all forms of videos, including what may be considered toxic, offensive, or indecent. Licensing: | [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/) ## Privacy Field | Response :----------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------- Generatable or reverse-engineerable personal information? | None known Protected class data used to create this model? | None known Was consent obtained for any personal data used? | None known How often is the dataset reviewed? | Before release Is a mechanism in place to honor data subject right of access or deletion of personal data? | Not applicable If personal data was collected for the development of the model, was it collected directly by NVIDIA? | Not applicable If personal data was collected for the development of the model by NVIDIA, do you maintain or have access to disclosures made to data subjects? | Not applicable If personal data was collected for the development of this AI model, was it minimized to only what was required? | Not applicable 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? | Not applicable ## Safety & Security Field | Response :---------------------------------------------------|:---------------------------------- Model Application(s): | World generation Describe the life-critical impact (if present). | None known Use Case Restrictions: | [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/) Model and dataset restrictions: | The Principle of least privilege (PoLP) is applied, limiting access for dataset generation and model development. Restrictions enforce dataset access during training, and ensures dataset license constraints are adhered to. Model checkpoints are made available on Hugging Face, and may become available on cloud providers' model catalogs. ## Prototype ```bash invoke_url='https://ai.api.nvidia.com/v1/cosmos/nvidia/cosmos-predict1-7b' fetch_url_format='https://api.nvcf.nvidia.com/v2/nvcf/pexec/status/' authorization_header='Authorization: Bearer $NVIDIA_API_KEY' accept_header='Accept: application/json' content_type_header='Content-Type: application/json' data='{ "inputs": [ { "name": "command", "shape": [1], "datatype": "BYTES", "data": [ "text2world --prompt=\"A first person view from the perspective from a human sized robot as it works in a chemical plant. The robot has many boxes and supplies nearby on the industrial shelves. The camera on moving forward, at a height of 1m above the floor. Photorealistic views\"" ] } ], "outputs": [ { "name": "status", "datatype": "BYTES", "shape": [1] } ] }' response=$(curl --silent -i -w "\n%{http_code}" --request POST \ --url "$invoke_url" \ --header "$authorization_header" \ --header "$accept_header" \ --header "$content_type_header" \ --data "$data" ) http_code=$(echo "$response" | tail -n 1) req_id=$(echo "$response" | grep -i '^nvcf-reqid:' | awk '{print $2}' | tr -d '\r') while [ "$http_code" -eq 202 ]; do response=$(curl --silent -i -w "\n%{http_code}" --request GET \ --url "$fetch_url_format$req_id" \ --header "$authorization_header" \ --header "$accept_header" \ --header "$content_type_header" \ ) http_code=$(echo "$response" | tail -n 1) req_id=$(echo "$response" | grep -i '^nvcf-reqid:' | awk '{print $2}' | tr -d '\r') done if [ "$http_code" -ne 302 ]; then echo "invocation failed with status $http_code" >&2 echo "$response" >&2 exit 1 fi download_url=$(echo "$response" | grep -i '^location:' | awk '{print $2}' | tr -d '\r') curl -L --output result.zip "$download_url" ``` ```javascript import fs from "fs"; const invokeUrl = "https://ai.api.nvidia.com/v1/cosmos/nvidia/cosmos-predict1-7b"; const fetchUrlFormat = "https://api.nvcf.nvidia.com/v2/nvcf/pexec/status/"; const headers = { "Authorization": "Bearer $NVIDIA_API_KEY", "Accept": "application/json", }; const payload = { "inputs": [ { "name": "command", "shape": [1], "datatype": "BYTES", "data": [ "text2world --prompt=\"A first person view from the perspective from a human sized robot as it works in a chemical plant. The robot has many boxes and supplies nearby on the industrial shelves. The camera on moving forward, at a height of 1m above the floor. Photorealistic views\"" ] } ], "outputs": [ { "name": "status", "datatype": "BYTES", "shape": [1] } ] }; let response = await fetch(invokeUrl, { method: "post", body: JSON.stringify(payload), headers: { "Content-Type": "application/json", ...headers } }); while (response.status === 202) { const requestId = response.headers.get("NVCF-REQID"); const fetchUrl = fetchUrlFormat + requestId; response = await fetch(fetchUrl, { method: "get", headers: headers }); } if (response.status !== 200) { const errBody = await (await response.blob()).text(); throw "invocation failed with status " + response.status + " " + errBody; } fs.writeFileSync('result.zip', Buffer.from(await response.arrayBuffer())); ``` ```python import requests invoke_url = "https://ai.api.nvidia.com/v1/cosmos/nvidia/cosmos-predict1-7b" fetch_url_format = "https://api.nvcf.nvidia.com/v2/nvcf/pexec/status/" headers = { "Authorization": "Bearer $NVIDIA_API_KEY", "Accept": "application/json", } payload = { "inputs": [ { "name": "command", "shape": [1], "datatype": "BYTES", "data": [ "text2world --prompt=\"A first person view from the perspective from a human sized robot as it works in a chemical plant. The robot has many boxes and supplies nearby on the industrial shelves. The camera on moving forward, at a height of 1m above the floor. Photorealistic views\"" ] } ], "outputs": [ { "name": "status", "datatype": "BYTES", "shape": [1] } ] } # re-use connections session = requests.Session() response = session.post(invoke_url, headers=headers, json=payload) while response.status_code == 202: request_id = response.headers.get("NVCF-REQID") fetch_url = fetch_url_format + request_id response = session.get(fetch_url, headers=headers) response = requests.post(invoke_url, headers=headers, json=payload) response.raise_for_status() with open('result.zip', 'wb') as f: f.write(response.content) ```