Version 1.16 (2026-05-15)
Supported Features
- More capable agents - New tools for graphs, embeddings, document parsing, markdown conversion, and table extraction; no hard cap on agent turns.
- Bird Cage execution - Replaces the previous sandbox with a faster, more reliable environment for running agent code.
- MCP integrations (beta) - Connect rooms and agents to external tools via MCP, with in-app setup for room-level integrations.
- Sharing conversation improvements - Revoke shared threads, working charts on shared links, and better previews for PDFs, CSVs, and large messages.
- Billing & usage - New usage dashboard, invoice listing, and current-cycle usage; fixes for trials and billing edge cases.
- Skills per agent - Enable or disable individual skills for each agent.
- Auth & security - Login via backend, WorkOS widget, finer-grained access tokens, dedicated user-content hosting, and ongoing security hardening.
- Agent v5 (internal only) - Redesigned agent UI with clearer progress, inline links and images in replies, and a smoother chat experience.
Version 1.15 (2026-04-24)
Supported Features
- Billing and plans - Stripe billing is enabled end-to-end, with clearer trials, seat-based pricing, and safer plan changes.
- More AI models - Anthropic Claude models are available, alongside an improved default model lineup.
- Smarter agents - More reliable answers after tool use, optional web search by default, and a refreshed “thought process” experience.
- Share conversations - Share chat threads with branded link previews and working charts.
- Easier onboarding - New org onboarding flow and fixes for sign-in and org creation.
- Behind the scenes - A new agent proxy layer improves how artifacts and data are served to the UI.
Version 1.14 (2026-04-15)
Supported Features
- Billing foundation - Developer plan added, Stripe enabled in dev, storage accounting for usage, and groundwork for end-to-end billing in the next release.
- Streaming thought process - Agent reasoning streams live in the UI, with fixes when multiple tables appear in the right panel.
- Feature view downloads - Download links for feature views directly in the app.
- Smarter table previews - Agents can preview a single row from very large tables without loading the full dataset.
- Clearer agent output - Tool results are grouped so you see the latest output; fixes for successful tool calls and large-table handling.
- Platform upgrades - Python 3.14, event log service (replacing the old event service), and reliability fixes for storage accounting and tracked tables.
- Behind the scenes - Frontend migrated from Poetry to uv; dependency and infrastructure updates across CI and workers.
Version 1.13 (2026-04-10)
Supported Features
- Billing service - New billing APIs, subscription cancellation, Stripe meter sync, and local billing when Stripe is off in an environment.
- Room & thread skills - Room-level skills with publish/demote to threads; updated skills, artifacts, and user-files UI.
- External agent tools - Configure which tools each agent can use, with UI for external tool integrations.
- Smarter agents - Per-message time limits, better web-search prompts, context-window fixes, GPT 5.4 support, and improved compose-response instructions.
- Canvas & charts - Auto-layout on the canvas, edge rendering fixes, and Vega charts upgraded to react-vega v8 with responsive layout.
- Data sources - Frontend wired to the new data-source service; file-upload fixes for resource tables.
- Downloads & previews - Feature view and file download URLs; non-UTF-8 CSV handling.
- Chat threads - Patch chat threads API; garbage collection for soft-deleted apps, threads, and tables.
- Platform & reliability - A2A agent HTTP surface, vector-search improvements, multi-token-issuer auth, and infrastructure monitoring on AWS/Azure.
Version 1.12 (2026-03-27)
Supported Features
- Skills - Agents can use skills (read, write, and edit files, view files, upload skills) with a new skills UI and per-thread skill handling.
- Web search & enrichment - Web search tool for agents, web sources in the app, and Web Augment for data enrichment via Parallel AI.
- Agent v3 - V3 is the default experience; artifact listing in the side panel, query interrupt, and streaming and markdown fixes.
- MCP for agents - MCP support for v3 agents, with MCP URLs shown in the agent integrations tab.
- Stripe integration - Initial Stripe billing integration on the backend (foundation for later billing releases).
- Smarter agents - Fixes for truncated file writes, compose-response streaming, Vega plots, Python tool without input tables, and faster interrupt handling.
- Chat improvements - Preview context files in threads, delete messages, and list artifacts from the side panel.
- Data apps & canvas - Data apps reorganized; node outputs refresh after runs; feature view download and column casing fixes.
- Auth - Simple-auth provider with code-grant login flow; health check endpoint.
- Team settings - Change member roles from the settings page; “Profile” renamed to “Settings”.
- Platform cleanup - Quick mode and visualization services removed; Poetry → uv migration; security headers on HTTP gateways.
Version 1.11 (2026-03-16)
Supported Features
- Agent v3 & canvas (beta) - Major V3 rollout: new canvas with auto-layout, Graph Builder, Join, Augment, and Embed operations; output detail pages; rename outputs; and broad UI polish.
- Agent Skills - Backend skills tools (read, write, edit, view files) and skills UI on the frontend, ahead of full skills enablement in the next release.
- Chat & context files - Upload files to chat threads, preview context files in chat, process-context-file tool, and notifications when file processing completes.
- Agent capabilities - Citations in Agent v3, query interrupt, exposed tool names/arguments, table output metadata with URLs, and Gemini 3.1 on the frontend.
- Agents & threads - Add, clone, and delete agents; threads on agents; connect agents to data apps; agent settings and metadata fixes.
- Data apps & sources - Document mode, directory/file data-source formats, error tables for failed ingestions, and canvas/data-source bug fixes.
- Auth & admin - Auth0 removed in favor of WorkOS; audit log tab and WorkOS audit export; change member roles; orgs without names supported in backoffice.
- Platform - Billing/management service split, Python 3.13, LiteLLM router with caching, FastAPI monolith support, and infrastructure gateway and monitoring updates.
Version 1.10 (2026-02-24)
Supported Features
- Database ingestion - Ingest database tables (with Airbyte worker support on the backend).
- Snowflake connector - Snowflake data connector in the UI.
- Cloud ingestion - Enabled everywhere; MCP support for deployed agents and cloud-ingestion syncing.
- Citations - Citation tags beside images in agent replies and broader citation rendering; feature flag removed for deployed agents.
- Agents & data apps - Agent metadata UI with auto-save, app delete/rename, and data-app list on room pages.
- Auth & backoffice - Auth0 removed; WorkOS APIs moved to org service; backoffice member panel and invite member/admin flows.
- Audit & admin - Audit reporter workflow, InviteAdmin RPC, and IAP-protected backoffice gateway.
- Platform - LiteLLM in agents, Deno sandbox with expanded Python/scientific libs, org/management service split, and infrastructure updates (Next.js 16, Python 3.13).
- Corvic V3 UI (internal only) - New V3 design end-to-end, with canvas nodes and edges, improved styling, and streaming agent responses.
- Agent v3 tools (internal only) - Python execution, Vega charts, read images, SQL across tables, semantic search/embed, relational-space tools, compose-response, and multi-tool-call support.
Version 1.9 (2026-01-28)
Supported Features
- Snap Actions UI - Snap Action and Snap Feature View flows in the app, with routers, error tables on actions, and clearer distinction between normal and snap actions.
- MCP for deployed agents - Agent ingress for MCP endpoints, MCP syncing quota, and fixes for agent sync and republish behavior.
- Sandbox in production - Sandbox service deployed on Azure and GCP for isolated code execution.
- Feature views - Error states on feature views, images rendered in table cells, loading states on details, and move from FeatureViewHead to FeatureViewUrls.
- Agent experience - Auto-scroll in reasoning, open context from any node in the action flowchart, improved reflection UI, and better markdown rendering for deployed agents.
- Data apps & canvas - Patch canvas locations, events when feature views are deleted, CSV support in structured and unstructured sources, and explore-data and graph display fixes.
- LLM Augment - Markdown columns treated as file prompts, clearer failure reasons, and duplicate output-name prevention.
- Platform - Gemini thinking level, local access-token creation, and documentation links pointed to docs.corvic.ai.
Version 1.8 (2026-01-08)
Supported Features
- Snap Actions - New Snap Actions capability for faster, guided workflows on data apps.
- Sandbox execution - Dedicated sandbox service for code augment and safer, isolated code execution.
- MCP for agents - Deploy agents as MCP endpoints, sync cloud ingestions and data apps, and check whether an agent has digested resources.
- Smarter orchestration - Tool-calling orchestrator enabled across deployments, with improved reflection prompts and action-parameter handling.
- Agent reasoning UI - Action details shown in agent reasoning; LLM Augment respects the configured model.
- Data apps - Rename feature views, optional output names on operations, and fixes for canvas nodes, citations, and extract-images after markdown conversion.
- Ingestion & storage - Storage quota checks on blob ingestion, document titles auto-generated when missing, and ingestion work moved off RPC hot paths.
- Auth cleanup - Auth0 variables removed; WorkOS-only auth path; file downloads/previews no longer routed through proxy URLs.
- Reliability - Agent metering fixes on delete, faster quota checks, Vega v6 upgrade, and broad lint and stability fixes across the UI.
Version 1.7 (2025-11-24)
Supported Features
- Introducing Data Apps! A new way to build workflows, explore data and chat with agents.
Version 1.6 (2025-10-17)
Supported Features
- Reflection mode now supported in agents.
Version 1.5 (2025-09-24)
Supported Features
- Citations in deployed agents
- Quota and plans
Version 1.4.3 (2025-09-19)
Supported Features
- Streamable HTTP for MCP
Version 1.4.2 (2025-09-02)
Supported Features
- Bug Fixes
- Fixed an issue ingesting parquet files
- Fixed an issue ingesting PDFs with many images on a page
Version 1.4.1 (2025-08-27)
Supported Features
- Bug Fixes
- Fixed an issue where image descriptions did not match their images
Version 1.4 (2025-08-16)
Supported Features
Multi-modal and Multi-structural Ingestion Pipelines:- Faster processing speeds with support for larger PDF files!
- Users can now incrementally add data from external data connectors like Amazon S3 and Azure Blob Storage.
- Users can upload CSV data to unstructured pipelines.
- Tabular embedding spaces now accept user input columns.
- Plots in agent messages can now be customised!
- Additional operations like DataAnalysis and FindRelevantDocuments have been added to support CSV data.
Evaluation Phase Limitations
Supported File Format: Unstructured Data:- PPTX
- MD
- TXT
- HTML
- DOCX
- CSV
- File Size: 10 MB
- Page Count: 200 pages
- Row Limit: 10 million rows per table
- For unstructured data (PDFs): Upload up to 5 files per batch for smoother and faster processing.
- For structured data (tables): Upload one table at a time to ensure proper ingestion.
- Unstructured Data (PDFs): If a single PDF exceeds 10 MB or 200 pages, split it into multiple PDFs.
- Split documents into 200 page PDFs to improve parsing times.
- Structured Data (Tables): If a table exceeds 10 million rows, split it into smaller tables before uploading.
- Ingestion Pipeline Limit: 1 GiB
- Total Room Limit: 5 GiB
- Data should be clean and ready to be used for feature engineering.
- Unique primary keys are expected in data for accurate data retrieval.
- New organization users should be added through Corvic support.
- Users need to configure primary and foreign keys for each source schema.
Version 1.3 (2025-07-17)
Supported Features
Multi-modal and Multi-structural Ingestion Pipelines:- Users can now upload data from external data connectors like Amazon S3 and Azure Blob Storage.
- Data Connectors:
- Admins can now configure external data sources like Amazon S3 buckets and Azure Blob Storage.
Version 1.2 (2025-07-09)
Supported Features
Multi-modal and Multi-structural Ingestion Pipelines:- Users can now upload Text, Markdown, HTML and Docx files in addition to pdf/pptx files for unstructured pipelines
- New Image embedding model supported:
- We have added support for SigLIP2, a multilingual vision-language encoder by Google DeepMind Research.
- To enhance customization of agents, we’ve introduced two new prompt types for the agent:
- Persona Prompt: Assign a distinct persona to the agent to customize its tone and style for more personalized interactions.
- Workflow Prompt: Introduce workflow-specific prompts that guide the agent with targeted instructions tailored to particular tasks or processes.
- New operations added to the Chain of Actions!
- Retrieve Adjacent: This operation performs one-hop neighbor retrieval, returning only the immediate (direct) connections of the specified input entities. This is useful to gather more precise context.
- Temporal Analysis: This operation analyzes time series data to identify underlying trends and seasonality. Seasonality Supported Features recurring patterns (e.g., sales peaks during holidays), while trend analysis reveals the overall direction of the data over time.
- SQL Queries now available in Agent Context.
- AI generated starter queries available for deployed agents.
- New look for the Integrations and Sharing Modal.
- Bug Fixes
- Fixed an issue where space generation would fail when using GCP models.
- Fixed a UI issue where the Unpublished Changes badge appeared after deployment, even when no changes were made.
- Fixed an issue where the integration and sharing modal displayed an incorrect MCP URL.
Version 1.1 (2025-06-11)
Supported Features
- Bug Fixes
- Fixed an issue with deployed agent being locked before being deployed.
- Disabled deployment preview when updating agent configuration to avoid race conditions.
- Fixed an issue where broken images and links were available in agent messages
- Update documentation with new section on MCP integrations
- Added support for citations of images on preview
Version 1.0 (2025-05-23)
Supported Features
Ingestion Pipelines:- Search and filter by status through multiple PDF files uploaded in a single pipeline.
- Cloning space configuration available now!
- Support for custom space names now available.
- Deployed agents now available as a Beta feature!
- Integrate MCP Endpoint to seamlessly unlock the power of Corvic agents in your system.
- Agents Citations is also available as a Beta feature!
- Agent answers now include references to source documents to support their responses.
Beta 2.4 (2025-05-18)
Supported Features
Multi-modal and Multi-structural Ingestion Pipelines:- Users can now upload pptx files in addition to pdf files for unstructured pipelines.
- Advanced user management:
- Admins can now organize and invite new members to their organization- saving time on setup and coordination.
- Enhanced Sign-in experience
- It’s now easier and more secure to manage who has access to your Corvic environment.
- More accurate AI insights
- Cloning agent configuration available now!
- Improved UI experience on Tablets and iPads
Beta 2.3 (2025-04-14)
Supported Features
Multi-modal and Multi-structural Ingestion Pipelines:- File download and preview options available now!
- Improved tracking for file uploads and failures.
- Custom LLM Integrations supported:
- Bring your own endpoints for agents to enable seamless data integration.
- Compatible API types:
- OpenAI Generic - Works with: Groq, Huggingface, Keyword AI, LM Studio, Ollama, OpenRouter, Together AI, Unify AI, vLLM
- OpenAI Azure - Works with: Azure
- Subscription plans:
- Corvic now offers a Premium and Enterprise tier. Reach out to support@corvic.ai for more information.
- Usage dashboard
- Track your data and computation usage across rooms
- Live context supported for Agents for incremental data ingestion.
- New custom query suggestions based on spaces available.
- New completion LLMs Supported:
- Custom LLM Integration endpoints
- Google: Gemini 2.0 Flash Thinking Preview
- Google: Gemini 2.0 Flash
Beta 2.1 (2025-03-10)
Supported Features
Multi-modal and Multi-structural Ingestion Pipelines:- Unstructured data (PDF): Documents, Images, Slides
- Structured data (Parquet): Tabular data
- Handles complex data transformations:
- PDF extraction including text, images, tables, graph layout
- Multi-file uploads
- Supports Incremental Ingestion
- Semantic embeddings supported:
- Google: GCP Vertex AI text-embedding-004
- OpenAI: Text-embedding-3-small
- OpenAI: Text-embedding-3-large
- Multi-model (Image) embeddings supported:
- Gemini: CLIP
- Tabular embeddings supported:
- Corvic: Categorical/Numerical Embed
- Graph embeddings supported:
- Corvic: Graph Structure Embed
- Multi-modal exploration
- Search images and text together
- Advanced search filters
- Easy export of multi-modal embeddings
- Multi-space agents
- Multi-modal chat: Retrieve and analyze images in chat
- Easy agent configuration
- Dynamic chain of actions
- Multi-modal context with explainability
- Major completion LLMs Supported:
- OpenAI: GPT 4o
- OpenAI: GPT 3.5 Turbo
- OpenAI: GPT 4o Mini
- Google: Gemini 2.0 Flash Preview
- Google: Gemini 1.5 Pro Preview
- Google: Gemini 1.5 Pro
- Google: Gemini 1.5 Flash Preview
- Google: Gemini 1.5 Flash
- Quick Mode functionality for automated workflow setup
Alpha 0.2 (2025-02-27)
Supported Features
- Vector search functionality with back-linking capabilities to original data
- Access control
- Interactive visualizations
- Improved structural transformation engine to support text-to-graph, table-to-text, and graph-to-text data conversions
- Out-of-the-box LLM offerings
Alpha 0.1
Supported Features
- Generation of embedding pipelines through the UI
- Uploading tabular data in the Parquet file format
- Autonomous table-to-graph data structure transformation
- Node2Vec graph embedding, utilizing graph transformation of Corvic Tables
- Visualized analysis of embeddings via UMAP plot
- Static embedding quality metrics: Stable Rank, Reciprocal Condition Number, and NE Sum.
- Embeddings export via Parquet format file download.
Coming Soon
Features
- Integration with cloud object storage services for data origins
- Autonomous schema extraction from data ingestion
- Advanced operations for agent policy generation
- Advanced quality metrics
- Python SDK and API-level integration
- Agent Deployment
- More Completion LLMs:
- Anthropic: Claude 3.5 Sonnet
- Anthropic: Claude 3 Haiku
- Mistral: Mistral Large