Complete Guide to Dreamina CLI: Practical Usage, Permission Activation & Common Problem Solving

Dreamina CLI (Command Line Interface) is an advanced operation method of the Dreamina platform, providing users with a more flexible path for AI-generated content operations. Especially in the scenario of All-Round Reference (Multimodal to Video), it can integrate images, videos and audio materials through the command line to generate dynamic videos that meet customized needs. However, in actual use, users generally encounter problems such as permission activation, function operation, and parameter understanding, such as account returning Error 3018, unclear image reference logic, and membership permission restrictions. Based on the core information of real user communication, this article comprehensively disassembles the usage method, permission activation process and common problem solutions of Dreamina CLI, helping users with college degree or above clearly grasp its usage logic and practical points.

1. Practical Operation of Dreamina CLI All-Round Reference Function: From Script to Implementation

The All-Round Reference (multimodal2video) usage script shared by user “Feng Ge” is a typical case of the core application scenario of Dreamina CLI. We will disassemble the script logic, parameter meaning and operation steps step by step, so that users can understand and adjust it according to their needs.

1.1 Core Logic Sorting of the Script

The core goal of the script is to convert static images into dynamic videos through the multimodal2video command of Dreamina CLI, combining the camera movement mode of the reference video and the music rhythm. The overall process is divided into 4 core links:

1. Material file existence check (pre-check to avoid task failure caused by missing files);
2. Submission of multimodal generation tasks (core operation, transmitting materials and generation parameters);
3. Task result analysis (extract task ID and generation status to master the progress);
4. Download of generation results (save the video to the specified directory after the task is successful).

1.2 Detailed Explanation of Script Parameters and Configurations

The following table clearly disassembles the meaning of core configuration items and command parameters in the script, helping users understand the role of each item:

1.3 Step-by-Step Operation Guide

Step 1: Configure File Paths (Basic Preparation)

First, modify the configuration items such as INPUT_DIR, OUTPUT_DIR, IMAGE, REF_VIDEO and BGM in the script, and replace them with the absolute paths of your own local materials. For example, if the materials are stored in D:/DreaminaMaterials, change INPUT_DIR to D:/DreaminaMaterials to ensure the path points to the file accurately, which is the key to avoiding the “file does not exist” error.

Step 2: File Existence Check (Pre-Check)

The script only checks the existence of the main image file by default, and the code logic is as follows:

if [ ! -f "$IMAGE" ]; then
echo "Error: Image does not exist $IMAGE"
exit 1
fi

The function of this step is to avoid material missing problems in advance. If the main image does not exist, the script will directly report an error and terminate operation. It is recommended that users supplement the existence check of REF_VIDEO and BGM as needed, for example:

if [ ! -f "$REF_VIDEO" ]; then
echo "Error: Reference video does not exist $REF_VIDEO"
exit 1
fi
if [ ! -f "$BGM" ]; then
echo "Error: Background music does not exist $BGM"
exit 1
fi

Step 3: Submit Multimodal Task (Core Operation)

Execute the dreamina multimodal2video command to submit the generation task. The complete command is disassembled as follows (line breaks are only for easy reading):

dreamina multimodal2video \
--image "$IMAGE" \
--video "$REF_VIDEO" \
--audio "$BGM" \
--prompt="Convert static images to dynamic videos according to the camera movement mode of the reference video and matching the music rhythm" \
--model_version=seedance2.0 \
--duration=8 \
--ratio=16:9 \
--poll=180 > result.json

After the command is executed, the task submission result will be output to the result.json file, including key information such as submit_id (task ID) and gen_status (generation status), which is the core basis for subsequent operations.

Step 4: Parse Task Results (Status Judgment)

The script extracts the task ID and status from result.json through the following commands:

SUBMIT_ID=$(cat result.json | grep -o '"submit_id": "\[^"\]*' | cut -d'"' -f4)
STATUS=$(cat result.json | grep -o '"gen_status": "\[^"\]*' | cut -d'"' -f4)

After extraction, “Task ID: XXX” and “Status: XXX” will be automatically output. Even if the status does not show “success”, you can query the progress later through the task ID.

Step 5: Download Generation Results (Operation After Success)

If STATUS is success, the script will automatically create the output directory and download the results:

if [ "$STATUS" = "success" ]; then
echo "Downloading..."
mkdir -p "$OUTPUT_DIR"
dreamina query_result \
--submit_id="$SUBMIT_ID" \
--download_dir="$OUTPUT_DIR"
echo "Completed! Output directory: $OUTPUT_DIR"
else
echo "Task not completed, please query later: dreamina query_result --submit_id=$SUBMIT_ID"
fi

Users can view the generated video files in the specified OUTPUT_DIR directory.

1.4 Script Optimization Suggestions

Users can adjust the script according to their own needs to improve practicality:

• Supplement full material check: add existence check of reference video and background music in addition to the main image;
• Optimize prompts: modify the --prompt content according to the scene, such as “Convert ancient-style images to dynamic videos with slow-motion camera movement of the reference video and light music rhythm”;
• Add retry logic: if the task status is not successful, add timed retry query logic to avoid manual repeated operations;
• Adjust generation parameters: modify --duration (duration) and --ratio (ratio) to adapt to different scenarios such as short videos and landscape videos.

2. Frequently Asked Questions (FAQ) about Dreamina CLI Usage

Based on the core questions of actual user communication, we divide the questions into four categories: Permission Activation, Function Usage, Membership & Points, and Experience & Limitations, and answer them one by one.

2.1 Permission Activation Related Questions

Q1: What should I do if I get Error 3018 (permission denied) when using Dreamina CLI?

A1: The core cause of Error 3018 is that the account is not on the whitelist, which is a permission restriction during the internal test of Dreamina CLI. The solution steps are as follows:

1. Collect key information: organize error information such as ret=3018, msg=permission denied, logId (e.g. 202603272023528BAAC406D31F299342A4);
2. Submit activation application: send the above information to Dreamina’s support/R&D personnel to apply for whitelist addition;
3. Wait for full activation: the official clearly stated that “the function is still in internal test and will be fully activated soon”, no additional membership subscription is required, just wait for permission opening.

Q2: How to apply for the usage permission of Dreamina CLI? What qualifications are required?

A2: Currently, the official does not require specific qualifications. The core is to submit information such as logId related to Error 3018 to Dreamina official personnel to apply for the whitelist. Some users asked if membership is required, and there is no evidence that membership subscription is required at present, only whitelist review is needed.

Q3: Where is the entrance for internal test application?

A3: No clear internal test application entrance is mentioned in user communication. The core method is still to submit information related to Error 3018 to Dreamina support/R&D personnel, or wait for the official full activation notice.

2.2 Function Usage Related Questions

Q1: In the All-Round Reference function, why can’t I @ uploaded images and quote specific images in the prompt like the WEB side?

A1: First, clarify the core logic: the @ image reference on the WEB side is essentially the same as directly uploading images without @, so directly uploading images in CLI can be recognized and used by the AGENT; second, if you need to quote different images in different shots (to avoid role confusion), you can solve it by optimizing prompts at present – let the AGENT describe the role characteristics in detail every time it uses images (such as “The ancient-style female character in Image 1, wearing a red long dress, as the main body in Shot 1”); the official also responded that “the function will be updated later to optimize the accuracy of image reference”.

Q2: In CLI, does mentioning “Image 1” and “Image 2” have the same effect as @ images on the WEB side? Is it recognized according to the upload order?

A2: Combined with the equivalence of @ images and direct upload on the WEB side, marking “Image 1” and “Image 2” in the upload order and describing them in detail in the prompt can achieve an effect similar to @ images on the WEB side. The core is to clarify the usage scenario of each image through the prompt instead of relying on the @ symbol.

Q3: How to generate four images at once? Do I need to let the AGENT perform four operations?

A3: No command or parameter for batch generating four images in CLI is mentioned in user communication. The current feasible method is to let the AGENT execute the generation operation four times separately, and there is no more efficient batch generation scheme for the time being.

Q4: Where can I view the model parameters of Dreamina CLI?

A4: Users reported that they could not find the complete model parameter description. Currently, only the model version parameter --model_version=seedance2.0 is seen in the practical script. The official has not disclosed more model parameter details for the time being, and you need to pay attention to subsequent updates.

Q5: Is Dreamina CLI suitable for pure beginners?

A5: There is a certain threshold for pure beginners to use. Some users reported that “directly copying the command to the AGENT resulted in failure”, the core reason is that they did not understand the script configuration logic (such as local absolute path modification, parameter meaning). It is recommended that beginners first disassemble the core steps of the script (file check, task submission, result analysis) before gradually modifying the configuration, instead of directly copying and using it.

2.3 Membership & Points Related Questions

Q1: Do I need a premium membership to generate images with Dreamina CLI? Why can’t standard members use it?

A1: According to actual user tests, standard members failed to generate images with CLI, while premium members can use it normally; at the same time, it should be noted that image generation with CLI uses the agent mode instead of the ordinary image generation mode. Users who originally had free points for image generation will be deducted points in CLI mode, which is caused by the difference in point rules between modes.

Q2: Is there a risk of malicious point consumption for premium membership accounts?

A2: There is a clear risk. Some users reported that “the premium membership account was defrauded of login information by others, and the other party used the open claw (Crayfish) tool to swipe points in the middle of the night”. Therefore, you must strictly protect the account login information and never disclose it to unknown personnel to avoid malicious consumption of points.

2.4 Experience & Limitation Related Questions

Q1: Do I need to queue to generate content with Dreamina CLI?

A1: Some users asked about queuing, but no clear queuing feedback is mentioned in the communication. It is speculated that there is no obvious queuing situation temporarily due to the small number of users during the internal test; there may be queuing requirements after full activation, subject to the actual official rules.

Q2: Is there a limit to the number of concurrent tasks of Dreamina CLI?

A2: Users asked “Is there a limit to the number of concurrent tasks at the same time?”, but there is no official clear limit on the number of concurrent tasks for the time being. It is recommended that users test gradually to avoid failure caused by submitting too many tasks at one time.

Q3: Can a daily token limit be added to Dreamina CLI and the web version?

A3: Some users suggested adding a daily token limit to prevent the AGENT from generating infinitely (similar functions are available on other platforms). Official feedback: users can implement this limit by themselves through CLI combined with vibecoding, and there is no official built-in token limit function for the time being.

Q4: Is there a time limit for the usage permission of Dreamina CLI? For example, can it only be used until May 1st?

A4: Some users asked this question, but no clear time limit is mentioned in the communication, subject to the subsequent official notice.

3. Precautions and Risk Tips for Dreamina CLI Usage

3.1 Technical Usage Precautions

1. Path specification: All material paths must use local absolute paths (such as D:/DreaminaMaterials/MainImage.png) to avoid file recognition failure caused by relative paths;
2. Parameter verification: Before submitting a task, make sure that the files corresponding to parameters such as --image, --video and --audio exist, and you can supplement a full check in the script;
3. Task query: If the status is not successful after the task is submitted, you can execute the command dreamina query_result --submit_id=XXX through submit_id to query later without repeated submission;
4. Prompt optimization: In multi-image reference scenarios, describe the roles/scenarios corresponding to the images in detail in --prompt to avoid role confusion.

3.2 Account & Security Risk Tips

1. Account protection: Never disclose the login information of premium membership accounts to others to prevent malicious consumption of points;
2. Permission application: Only submit activation applications to Dreamina official support/R&D personnel, and avoid providing information such as logId to unknown personnel;
3. Point consumption: Clarify the point rules for image generation with CLI (points are deducted in agent mode) to avoid excessive point consumption due to unknown rules.

4. Summary of User Feedback & Official Responses

From the overall feedback of user communication, the core value of Dreamina CLI is to provide users with a technical foundation a flexible operation method for AI-generated content, especially in the multimodal to video scenario, which can precisely control generation parameters through scripts; however, permission restrictions during the internal test (Error 3018, whitelist), usage threshold for beginners, membership and point rules are still the core issues concerned by users.

The core official responses can be summarized into three points:

1. Permission activation: Users need to submit information related to Error 3018 to apply for the whitelist, and the function will be fully activated later;
2. Function optimization: The image reference function will be updated to solve the accuracy of quoting different images in different shots;
3. Personalized needs: Such as token limit, can be implemented by users through CLI combined with vibecoding.

Conclusion

As an advanced tool of the Dreamina platform, Dreamina CLI has significant flexibility advantages in the scenario of multimodal content generation, but problems such as permission restrictions and usage thresholds during the internal test still need to be solved gradually. For users with college degree or above, mastering the core script logic, parameter configuration and problem troubleshooting methods of CLI can better use Dreamina’s AI capabilities to generate customized content; at the same time, protecting account security and paying attention to official updates can further improve the usage experience. With the full activation and optimization of functions, Dreamina CLI is expected to become an efficient tool in the field of AI content generation, adapting to more personalized generation needs.

Would you like me to add Google SEO meta tags and schema markup for this English blog to further optimize search engine indexing?