Skip to main content

Release 10.0

Release Date: January 9th, 2024

New FeatureImprovementBug FixEnterprise Only

Text Generation

new-featureIntroduced a UI for text generation modelsThe Model-Viewer screen of text generation models now has a revamped UI that lets you effortlessly generate or convert text based on a given text input prompt.

  • For third-party wrapped models, like those provided by OpenAI, you can choose to utilize their API keys as an option, in addition to using the default Clarifai keys.
  • Optionally, you can enrich the model's understanding by providing a system prompt, also known as context.
  • Optionally, inference parameters are available for configuration. They are hidden by default.
  • The revamped UI provides users with versatile options to manage the generated output. You can regenerate, copy, and share output.
new-featureAdded more training templates for text-to-text generative tasks
  • You can now use Llama2 7/13B and Mistral templates as a foundation for fine-tuning text-to-text models.
  • There are also additional configuration options, allowing for more nuanced control over the training process. Notably, the inclusion of quantization parameters via GPTQ enhances the fine-tuning process.


new-featureIntroduced the RAG-Prompter operator model
new-featureImproved the process of making predictions on the Model-Viewer screenTo make a prediction using a model, navigate to the model’s viewer screen and click the Try your own input button. A modal will pop up, providing a convenient interface for adding input data and examining predictions.

The modal now provides you with three distinct options for making predictions:

  • Batch Predict on App Inputs—allows you to select an app and a dataset. Subsequently, you’ll be redirected to the Input-Viewer screen with the default mode set to Predict, allowing you to see the predictions on inputs based on your selections.
  • Try Uploading an Input—allows you to add an input and see its predictions without leaving the Model-Viewer screen.
  • Add Public Preview Examples—allows model owners to add public preview examples.
improvementImproved the UI/UX of the models’ evaluation leaderboard
  • Replaced the Context based classifier wording with Transfer learn.
  • Added a dataset filter functionality that only lists datasets that were successfully evaluated.
  • A full URL is now displayed when hovering over the table cells.
  • Replaced "-" of empty table cells in training and evaluation dataset columns with "{{appName}}-all-app-inputs"
improvementAdded support for inference settings
  • All models have undergone updates to incorporate new versions that now support inference hyperparameters like temperature, top_k, etc. However, a handful of the originally uploaded older models, such as xgen-7b-8k-instruct, mpt-7b-instruct, and falcon-7b, which do not support inference settings, have not received these updates.
bugFixed an issue where detection and segmentation models generated the wrong code
  • Previously, if you clicked the “Use Model” button, the code generated by detection models, such as General Detection and Face Detection, and segmentation models, such as General Segmentation, could only work with classification models. We fixed the issue.
bugFixed an issue with the model version dropdown
  • Previously, the dropdown on the model's page displayed the latest model version, irrespective of whether that version was marked as trained and ready for use in the versions table. We’ve fixed the issue, and the dropdown now accurately reflects the trained and ready-to-use model versions.
bugFixed an issue that prevented model training with default settings
  • Previously, attempts to train certain models, particularly visual classifiers and visual detectors, using the recommended default settings resulted in numerous errors. We fixed the issue.

New Published Models

new-featurePublished several new, ground-breaking models
  • Wrapped Fuyu-8B, an open-source, simplified multimodal architecture with a decoder-only transformer, supporting arbitrary image resolutions, and excelling in diverse applications, including question answering and complex visual understanding.
  • Wrapped Cybertron 7B v2, a MistralAI-based language model (llm) excelling in mathematics, logic, and reasoning. It consistently ranks #1 in its category on the HF LeaderBoard, enhanced by the innovative Unified Neural Alignment (UNA) technique.
  • Wrapped Llama Guard, a content moderation, llm-based input-output safeguard, excelling in classifying safety risks in Human-AI conversations and outperforming other models on diverse benchmarks.
  • Wrapped StripedHyena-Nous-7B, an innovative hybrid chat llm, featuring multi-head attention and gated convolutions, outperforms Transformers in long-context summarization with notable efficiency improvements in training and inference.
  • Wrapped Imagen 2, a cutting-edge text-to-image llm, offering high-quality, multilingual image generation with advanced features, including improved text rendering, logo generation, and safety measures.
  • Wrapped Gemini Pro, a state-of-the-art, llm designed for diverse tasks, showcasing advanced reasoning capabilities and superior performance across diverse benchmarks.
  • Wrapped Mixtral 8x7B, a high-quality, Sparse Mixture-of-Experts (SMoE) llm model, excelling in efficiency, multilingual support, and competitive performance across diverse benchmarks.
  • Wrapped OpenChat-3.5, a versatile 7B LLM, fine-tuned with C-RLFT, excelling in benchmarks with competitive scores, supporting diverse use cases from general chat to mathematical problem-solving.
  • Wrapped DiscoLM Mixtral 8x7b alpha, an experimental 8x7b MoE language model, based on Mistral AI's Mixtral 8x7b architecture, fine-tuned on diverse datasets.
  • Wrapped SOLAR-10.7B-Instruct, a powerful 10.7 billion-parameter LLM with a unique depth up-scaling architecture, excelling in single-turn conversation tasks through advanced instruction fine-tuning methods.


new-featureIntroduced the Databricks-Connect UI module for integrating Clarifai with DatabricksYou can use the module to:

  • Authenticate a Databricks connection and connect with its compute clusters.
  • Export data and annotations from a Clarifai app into Databricks volume and table.
  • Import data from Databricks volume into Clarifai app and dataset.
  • Update annotation information within the chosen Delta table for the Clarifai app whenever annotations are getting updated.

Base Workflow

improvementChanged the default base workflow of a default first application
  • Previously, for new users who skipped the onboarding flow, a default first application was generated having "General" as the base workflow. We’ve replaced it with the "Universal" base workflow.

Python SDK

new-featureAdded support for Secure Data Hosting (SDH)
  • The SDK now supports the SDH feature for uploading and downloading user inputs.
improvementAdded vLLM template for model upload to the SDK
  • This additional template expands the range of available templates, providing users with a versatile toolset for seamless deployment of models within their SDK environments.
bugFixed an issue with the CocoDetection loader
  • Previously, the CocoDetection loader had a bug that caused it to convert bounding box coordinates from the xywh format (center coordinates, width, height) to the xyxy format (top-left, bottom-right coordinates) using the wrong divisor. Instead of dividing by the bounding box's height, it was incorrectly dividing by its width. We fixed it.


bugImproved pagination handling during multiple input deletion in the Input-Manager
  • Previously, there was an issue of pagination fetch inconsistencies after deleting multiple inputs or annotations. Now, when you delete a selection of inputs or annotations, the pagination mechanism resets to ensure a more accurate and streamlined retrieval of pages.
bugFixed an issue with performing visual search on the Input-Manager
  • Previously, if you performed a face similarity search in an app with "Face" as the base workflow, it could return no outputs. We fixed the issue.
bugFixed an issue where it was not possible to exit the Upload Inputs pop-up window
  • Previously, there was no way to exit the pop-up if there were no inputs to upload—users were forced to refresh the page. We fixed the issue.

Labeling Tasks

new-featureAdded ability to view and edit previously submitted inputs while working on a task
  • We have added an input carousel to the labeler screen that allows users to go back and review inputs after they have been submitted. This functionality provides a convenient mechanism to revisit and edit previously submitted labeled inputs.
bugFixed an issue with deleting a task on the Tasks page
  • Previously, deleting the first task in a list of multiple tasks did not work as expected. While the deletion request was processed, the row representing the deleted task remained visible, resulting in an unusual and weird state. We fixed the issue.
bugFixed an issue with using AI-Assist on the Labeler screen, where predictions were sometimes prone to becoming "stuck"
  • Previously, when labeling using the AI-Assist feature, and you tried to approve or reject an image by clicking the checkmark or the x button, an unintended consequence arose—multiple concept IDs were generated with each action. Subsequently, the bounding box annotations associated with these additional concept IDs persisted on the screen for the remaining images in the task, rendering them immovable. We fixed the issue.


improvementMade enhancements to the App Settings page
  • Added a new collaborators table component for improved functionality.
  • Improved the styling of icons in tables to enhance visual clarity and user experience.
  • Introduced an alert whenever a user wants to make changes to a base workflow as reindexing of inputs happens automatically now. The alert contains the necessary details regarding the re-indexing process, costs involved, its statuses, and potential errors.
improvementEnhanced the inputs count display on the App Overview page
  • The tooltip (?) now precisely indicates the available number of inputs in your app, presented in a comma-separated format for better readability, such as 4,567,890 instead of 4567890.
  • The display now accommodates large numbers without wrapping issues.
  • The suffix 'K' is only added to the count if the number exceeds 10,000.


improvementAdded “Last Updated” date in resources alterations
  • We’ve replaced "Date Created" with "Last Updated" in the sidebar of apps, models, workflows, and modules (these are called resources).
  • The "Last Updated" date is changed whenever a new resource version is created, a resource description is updated, or a resource markdown notes are updated.
improvementAdded a "No Starred Resources" screen for models/apps empty state
  • We’ve introduced a dedicated screen that communicates the absence of any starred models or apps in the current filter when none is present.
improvementEnhanced the resource overview page with larger images where possible
  • We now include rehosting of large versions of resource cover images alongside small ones. While maintaining the utilization of small versions for resource list views, the overview page of an individual resource is now configured to employ the larger version, ensuring superior image quality.
  • Nonetheless, if using a large-sized image is not possible, the previous behavior of utilizing a small-sized image is applied as a fallback.
improvementEnhanced image handling in listing view
  • Fixed an issue where cover images were not being correctly picked up in the listing view. Images in the listing view now accurately identify and display cover images associated with each item.
improvementEnhanced search queries by including dashes between text and numbers
  • For instance, if the search query is "llama70b" or "gpt4," we also consider "llama-70-b" or "gpt-4" in the search results. This provides a more comprehensive search experience.
improvementRevamped code snippets presentation
  • We’ve updated the code snippet theme to a darker and more visually appealing scheme.
  • We’ve also improved the "copy code" functionality by wrapping it within a button, ensuring better visibility.
bugFixed an issue with the "Back to Community" buttonWhen attempting to access an unavailable resource in the Clarifai Community portal, users are directed to a page indicating the non-existence of the resource. And a button is provided that users can click and return to the portal.

  • Previously, the button could not redirect users. Clicking the button now correctly redirects users back to the Community portal.
bugFixed an issue that prevented deleting a cover image
  • You can now remove a cover image from any resource—apps, models, workflows, datasets, and modules—without any problems.

Organization Settings and Management

improvementAllowed members of an organization to work with the Labeler Tasks functionality
  • The previous implementation of the Labeler Tasks functionality allowed users to add collaborators for working on tasks. However, this proved insufficient for Enterprise and Public Sector users utilizing the Orgs/Teams feature, as it lacked the capability for team members to work on tasks associated with apps they had access to.
  • We now allow admins, org contributors, and team contributors with app access to work with Labeler Tasks.


improvementDisabled the "Please Verify Your Email" popup
  • We deactivated the popup, as all accounts within on-premises deployments are already being automatically verified. Furthermore, email does not exist for on-premises deployments.