AutoGPT
AI & LLMsAutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
Release History
View all versions →autogpt-platform-beta-v0.7.36 fixes7 featuresThis release introduces a new email system design, enhances Copilot with selectable options and live progress, and includes several UI/UX improvements and bug fixes. It also recommends Ornith 1.5 for local users and fires Google Ads conversions across the signup-to-paid journey.
autogpt-platform-beta-v0.7.211 fixes19 featuresThis release introduces a wide range of new features including AI-voice narratives, per-expert spend tracking, and a Microsoft Teams bot adapter. It also includes numerous bug fixes and UI/UX improvements to enhance the user experience and system stability.
autogpt-platform-beta-v0.7.114 fixes16 featuresThis release introduces new features like Tavily provider blocks, Claude Sonnet 5 support, and enhanced expert scheduling. It also includes various bug fixes, UI/UX improvements, and developer experience enhancements, alongside a new single-container distribution option.
preview-seed-fixture3 featuresThis release introduces a new synthetic database seed fixture for preview environments, improving the development workflow by ensuring schema changes are tested against populated tables. The fixture is automatically regenerated and is designed to be disposable and secure.
autogpt-platform-beta-v0.7.06 fixes6 featuresThis release introduces significant new features including expert-scoped sessions, an experts marketplace, and a new LLM model catalog. It also includes several UI/UX improvements and bug fixes for a more robust user experience.
autogpt-platform-beta-v0.6.707 fixes5 featuresThis release introduces several new features including configurable transcription endpoints, image pasting into CoPilot chat, and enhanced library sorting. It also includes numerous bug fixes for improved stability and performance.
autogpt-platform-beta-v0.6.6915 fixes7 featuresThis release introduces a new app sidebar layout, proactive posting for Slack and Telegram, and an Agent-Building Mode. It also includes numerous bug fixes and enhancements to improve user experience and agent functionality.
autogpt-platform-beta-v0.6.683 fixes2 featuresThis release includes improvements to the product tour experience and updates to the public platform README. It also addresses several bug fixes related to webhook presets, OpenAPI schema regeneration, and binary bot file handling.
autogpt-platform-beta-v0.6.67Breaking14 fixes11 featuresThis release introduces significant new features including first-class organization/workspace support, new adapters for Slack and Telegram, and an enhanced Copilot composer. It also includes numerous bug fixes and dependency updates.
autogpt-platform-beta-v0.6.6611 fixes9 featuresThis release introduces significant new features like file uploads to AutoPilot via Discord and chat pinning, alongside crucial bug fixes related to data handling and security hardening in classic mode. It also includes a dependency update to Node 24 LTS.
autogpt-platform-beta-v0.6.659 fixes8 featuresThis release introduces significant new features like the AutoPilot Context Panel and global Cmd+K search, alongside numerous reliability fixes for the Copilot and LLM providers.
autogpt-platform-beta-v0.6.649 fixes8 featuresThis release introduces significant new features like the AutoPilot Context Panel and Global Search command palette, alongside numerous reliability and security hardening improvements across the platform.
autogpt-platform-beta-v0.6.6325 fixes13 featuresThis release introduces significant new features like public link sharing for chat results and native scheduling for Copilot followups. It also includes numerous bug fixes focused on credential handling, orchestrator stability, and UI polish.
autogpt-platform-beta-v0.6.6229 fixes13 featuresThis release introduces significant new features like public chat sharing, self-distilled skills, and native scheduling for Copilot. It also includes numerous bug fixes focused on credential handling, orchestrator stability, and UI polish.
autogpt-platform-beta-v0.6.617 fixes4 featuresThis release introduces new features like chat search and profile redesign, enhances session handling with pagination, and fixes several bugs related to credentials, onboarding, and UI hints. It also retires deprecated LLM models.
autogpt-platform-beta-v0.6.6021 fixes9 featuresThis release introduces significant enhancements to workflow triggers, Slack integration, and cost estimation, alongside numerous stability fixes and performance improvements, particularly around LLM routing via OpenRouter.
autogpt-platform-beta-v0.6.5915 fixes13 featuresThis release introduces significant platform enhancements, including new billing features like yearly Stripe billing and tier-based storage limits. It also includes numerous bug fixes focused on chat stability, payment processing, and user onboarding flow adjustments.
autogpt-platform-beta-v0.6.5817 fixes20 featuresThis release introduces significant platform upgrades, including the rollout of Settings v2 pages, dynamic cost billing models (COST_USD), and support for Claude Opus 4.7. Numerous stability improvements were made, particularly within Copilot's streaming and session management.
autogpt-platform-beta-v0.6.5715 fixes18 featuresThis release introduces significant new features including enhanced billing via Stripe, expanded model support (xAI Grok 4.20, Kimi K2.6), and major UI/UX improvements like the Agent Briefing Panel and flow builder chat. Numerous bug fixes address session management, pagination, and model defaults.
autogpt-platform-beta-v0.6.561 fix3 featuresThis release introduces memory hardening, scoped retrieval, and fixes an issue in the copilot backend related to message role assignment.
autogpt-platform-beta-v0.6.5517 fixes8 featuresThis release introduces significant enhancements to cost tracking, memory management via Graphiti, and improved session handling in CoPilot. It also includes numerous bug fixes related to SSE streams, session state persistence, and security hardening.
autogpt-platform-beta-v0.6.5431 fixes27 featuresRelease v0.6.54 introduces significant new features focused on CoPilot enhancements, agent execution modes, and credential management, alongside numerous bug fixes and maintenance updates including Codecov integration.
autogpt-platform-beta-v0.6.5311 fixes5 featuresThis release introduces significant new features like workflow import, parallel block execution, and dry-run mode. It also includes numerous bug fixes, dependency bumps, and documentation improvements, notably renaming SmartDecisionMakerBlock to OrchestratorBlock.
autogpt-platform-beta-v0.6.5219 fixes16 featuresThis release introduces significant new features like whitelisting for user provisioning, prompt-in-URL support, and GitHub CLI integration. It also includes numerous bug fixes, UI enhancements, and improvements to agent execution and testing infrastructure.
autogpt-platform-beta-v0.6.51Breaking27 fixes15 featuresThis release introduces significant new features like local agent generation, multimodal input support, and hardened GitHub blocks. It also expands LLM support with numerous new models and includes various bug fixes related to file handling, SDK integration, and UI stability.
Common Errors
BlockUnknownError22 reportsThis error typically occurs in autogpt. Check the example issues for common solutions.
BlockExecutionError5 reportsBlockExecutionError in AutoGPT usually indicates a problem during the execution of a specific block in the agent's workflow, often due to invalid input/output data or unexpected errors within the block's logic. Inspect the error message for the failing block's ID and reason, then address the root cause by validating data formats, handling potential exceptions in the block's code, or correcting any configuration issues specific to that block type. Thoroughly test the corrected workflow to ensure the error is resolved and the agent functions as expected.
ParamValidationError4 reportsParamValidationError in AutoGPT often indicates an issue with the format or validity of the input parameters being sent to an API or function, such as the URL in ExtractWebsiteContentBlock. To fix it, carefully validate your input parameters against the expected schema or requirements of the receiving API or function *before* sending the request, paying close attention to data types, allowed values, and required fields, and implement specific error handling around the parameter validation. You can use tools like pydantic to enforce these data constraints directly in your code.
AssertionFailureError2 reportsAssertionFailureError in AutoGPT often arises from unexpected API responses, such as empty search results or failed fact checks, violating expected conditions within a block. Implement robust error handling with try-except blocks and conditional checks to gracefully manage these responses, providing alternative actions or informative messages instead of crashing. Consider implementing retry mechanisms with exponential backoff for transient API failures.
RawQueryError1 reportRawQueryError in AutoGPT usually means the SQL query sent to the database is malformed or incompatible with the database schema. To fix it, carefully examine the SQL query in the error message (especially after any recent code changes), and ensure it correctly references existing tables and columns while adhering to the database's specific SQL syntax. Consider using a database management tool to manually validate the query and schema.
ForeignKeyViolationError1 reportForeignKeyViolationError means you're trying to create a database record with a foreign key value that doesn't exist in the related table. To fix this, ensure the referenced record exists (e.g., a user with the specified userId) *before* creating the new record, or create the corresponding record first. Verify that the data being passed for the foreign key is accurate and valid based on the existing entries in the referenced table.
Related AI & LLMs Packages
Get up and running with OpenAI gpt-oss, DeepSeek-R1, Gemma 3 and other models.
🦜🔗 The platform for reliable agents.
The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.
LLM inference in C/C++
GPT4All: Run Local LLMs on Any Device. Open-source and available for commercial use.
A high-throughput and memory-efficient inference and serving engine for LLMs
Subscribe to Updates
Get notified when new versions are released