v1.0.0

Morphic v1.0.0

A major release with significant architectural changes. This version is a complete rewrite of the core infrastructure.

Breaking Changes

• Database: Redis → PostgreSQL — Chat history storage has moved from Redis (Upstash) to PostgreSQL with Drizzle ORM. Previous chat history stored in Redis will not be migrated. No migration script is provided. Run bun run migrate to create the new database tables.
• Next.js 15 → 16 — Upgraded to Next.js 16 with Turbopack.
• AI SDK 4 → 6 — Upgraded to AI SDK 6 with ToolLoopAgent pattern.
• Environment variables — Several environment variables have been added, removed, or renamed. Refer to .env.local.example for the current required configuration.

Summary

Core

• AI-powered search with GenerativeUI
• Next.js 16 with App Router and Turbopack
• AI SDK 6 integration with ToolLoopAgent
• PostgreSQL with Drizzle ORM and Row-Level Security (RLS)
• Supabase authentication with guest mode support

AI Providers

• OpenAI (default), Anthropic Claude, Google Gemini, Vercel AI Gateway, Ollama
• Profile-based model configuration (config/models/*.json)

Search

• Multiple providers: Tavily, Brave, SearXNG, Exa, Firecrawl
• Search modes: Quick, Planning, and Adaptive
• URL-specific search and content extraction

Features

• File upload with Cloudflare R2
• Chat history with share functionality
• Message feedback system
• Inspector panel for tool execution details
• Todo tracking for complex tasks
• Changelog notification system
• LLM observability with Langfuse (optional)

Deployment

• Docker deployment with prebuilt images (anonymous mode only)
• Vercel deployment with one-click setup
• Docker Compose for self-hosted setup with PostgreSQL, Redis, and SearXNG

New Contributors

• @oussamadhouib made their first contribution in #556
• @mcmadafly made their first contribution in #631
• @dan-and made their first contribution in #708

Full Changelog: v0.4.6...v1.0.0

v1.0.0-beta.10

What's Changed

• feat: update model configuration and optimize tool selection by @miurla in #733
• feat: replace citation numbers with domain names by @miurla in #734
• fix: resolve chatId duplication issue in Next.js 16 by @miurla in #735
• refactor: clarify fetch tool PDF support and type usage by @miurla in #736
• fix: remove plus icon overlay from image grid by @miurla in #737
• fix: reload chat after message deletion in regeneration flow by @miurla in #738
• test: add comprehensive tests for prepareMessages regeneration flow by @miurla in #739
• feat(search): add source favicons to search results display by @miurla in #740
• feat: add parent collapsible for research sections with 5+ parts by @miurla in #741
• refactor: redesign citation popover layout by @miurla in #742
• feat: add explicit efficiency guidelines to research prompts by @miurla in #743
• feat: improve search mode prompts for flexible response length by @miurla in #745
• feat: integrate pruneMessages for optimized context window management by @miurla in #746
• fix(security): update Next.js to 16.0.7 for CVE-2025-66478 by @miurla in #749
• fix(security): update Next.js to 15.2.6 for CVE-2025-66478 by @miurla in #750
• security: update Next.js to 16.0.10 for additional CVE fixes by @miurla in #753
• security: update Next.js to 15.2.8 for CVE fixes by @miurla in #754
• fix: remove client-side auth check for Docker anonymous mode by @miurla in #757
• fix: improve auth error handling in ErrorModal by @miurla in #758
• feat: migrate to AI SDK v6 with ToolLoopAgent by @miurla in #760
• chore: update Next.js to 16.1.6 by @miurla in #761
• fix: handle null value for exclude_domains validation by @miurla in #762
• fix: keep reasoning in pruneMessages for OpenAI compatibility by @miurla in #763
• fix: strip reasoning parts for OpenAI models to fix multi-turn errors by @miurla in #765
• feat: add welcome message below logo on home screen by @miurla in #766
• Add guest mode and chat limits by @miurla in #767
• docs: add guest mode documentation by @miurla in #768
• docs: update README for v1.0.0 release by @miurla in #769

Full Changelog: v1.0.0-beta.9...v1.0.0-beta.10

v1.0.0-beta.9

What's Changed

• feat: integrate Ollama model discovery with tools filtering by @dan-and in #708
• chore: remove static Ollama placeholders by @miurla in #709
• fix: Fix header scrolling issue on iOS Safari by @miurla in #719
• feat: Restore Ollama provider support for v1.0.0 (#721) by @miurla in #722
• Fix beta badge visibility and mobile layout improvements by @miurla in #724
• Fix mobile chat message padding with accurate offset calculation by @miurla in #725
• feat: Upgrade to Next.js 16 with Turbopack and MCP support by @miurla in #726
• feat: Add Docker support with PostgreSQL, Redis, and optional authentication by @miurla in #727
• fix: Hide scroll-to-bottom button immediately on click by @miurla in #728
• feat: make BRAVE_SEARCH_API_KEY optional with dynamic provider fallback by @miurla in #729
• fix: Fix scroll listener registration to properly detect bottom position by @miurla in #730
• refactor: remove planning search mode and integrate into adaptive mode by @miurla in #731
• feat: add emoji usage guidelines to search mode prompts by @miurla in #732

New Contributors

• @dan-and made their first contribution in #708

Full Changelog: v1.0.0-beta.8...v1.0.0-beta.9

v1.0.0-beta.8

Beta Testing Now Available

Starting with this release, v1.0.0 is now available for beta testing at beta.morphic.sh.

We welcome your feedback as we work toward the stable release. Please report any issues or suggestions through GitHub Issues or use the feedback button in the app.

For more information about v1.0.0, see Issue #680.

---

What's Changed

• feat: implement model matrix and improve citation system by @miurla in #711
• feat: migrate researcher agent to streamText with smoothStream by @miurla in #712
• fix: restore cursor pointer for buttons in Tailwind v4 by @miurla in #713
• Keep single tool sections open and stabilize image previews by @miurla in #714
• feat: Add quality mode rate limiting with user-friendly error handling by @miurla in #715
• feat: Improve search error messages for better user experience by @miurla in #716
• feat: Add Vercel Analytics tracking for chat events by @miurla in #717

Full Changelog: v1.0.0-beta.7...v1.0.0-beta.8

v1.0.0-beta.7

What's Changed

• feat: start chat streaming optimistically by @miurla in #704
• refactor: streamline search image filtering by @miurla in #705
• Stream related questions incrementally by @miurla in #706
• Update changelog for v1.0.0 Beta.7 by @miurla in #707

Full Changelog: v1.0.0-beta.6...v1.0.0-beta.7

v1.0.0-beta.6

What's Changed

• fix: prevent displaying '0' results during loading in search section by @devin-ai-integration[bot] in #595
• feat(components): add password visibility toggle to Input by @vishv96 in #663
• ISSUE-546: by @Dhaval-x96 in #671
• feat(cache): implement selective caching for individual chat loading by @miurla in #690
• chore: update AI SDK dependencies to latest versions by @miurla in #691
• feat: implement streaming support for AI tools with enhanced type safety by @miurla in #692
• fix: resolve input focus ring persistence after submission by @miurla in #693
• feat: enhance AI prompt system with improved planning and search capabilities by @miurla in #694
• Fix enter handling in user text editor by @miurla in #696
• Firecrawl based Search Provider by @yazaldefilimone in #695
• Fix assistant message deletion under RLS by @miurla in #697
• Remove legacy message-preparation helper by @miurla in #699
• Reorder search modes and set quick as default by @miurla in #700
• Fix citation mapping for message copy by @miurla in #701
• Merge main into v1.0.0 by @miurla in #702

New Contributors

• @vishv96 made their first contribution in #663
• @Dhaval-x96 made their first contribution in #671
• @yazaldefilimone made their first contribution in #695

Full Changelog: v1.0.0-beta.5...v1.0.0-beta.6

v0.4.6

What's Changed

• Update README: Add Sponsors and License sections, remove Verified models by @miurla in #560
• Fix sidebar state persistence and hydration flash by @miurla in #563
• feat: Use image from search results for OG tag by @ngoiyaeric in #567
• Add CLAUDE.md documentation and remove .cursor directory by @miurla in #574
• Add typecheck, format commands and CI workflow by @miurla in #575
• fix: prevent displaying '0' results during loading in search section by @devin-ai-integration[bot] in #595
• feat(components): add password visibility toggle to Input by @vishv96 in #663
• ISSUE-546: by @Dhaval-x96 in #671
• Firecrawl based Search Provider by @yazaldefilimone in #695

New Contributors

• @ngoiyaeric made their first contribution in #567
• @vishv96 made their first contribution in #663
• @Dhaval-x96 made their first contribution in #671
• @yazaldefilimone made their first contribution in #695

Full Changelog: v0.4.5...v0.4.6

v1.0.0-beta.5

What's Changed

• Harden site feedback insert under RLS; add SELECT policy and Drizzle migration by @miurla in #681
• refactor(ui): replace react-markdown with streamdown and simplify code blocks (Refs #683) by @miurla in #684
• docs(prompt): enforce TODO completion via prompts (Refs #682) by @miurla in #685
• style(ui): compact Thoughts header in reasoning section by @miurla in #686
• ci(release): add tag-driven release workflow (pre-release gating by base version) by @miurla in #687
• feat(ui): add ResearchProcessSection with accordion, fix data rendering, and update FetchSection by @miurla in #688
• Release v1.0.0-beta.5 by @miurla in #689

Full Changelog: v1.0.0-beta.4...v1.0.0-beta.5

v1.0.0-beta.4

Note: Advancing from v0.5 to v1.0 as a major release due to substantial cumulative improvements.

What's Changed

• feat: Add Row-Level Security (RLS) to all database tables by @miurla in #656
• fix: Improve tool error handling for AI SDK v5 by @miurla in #658
• fix: stabilize message height and improve chat scroll position by @miurla in #659
• feat: Implement Row-Level Security (RLS) for multi-tenant data isolation by @miurla in #660
• fix: add RLS context for feedback submission by @miurla in #661
• fix: add RLS context for message feedback updates by @miurla in #662
• feat: Enhanced search modes with improved UI and model selection by @miurla in #664
• feat: optimize fetch tool limits and add quick mode restrictions by @miurla in #665
• Fix: Display message actions only on last text part by @miurla in #666
• refactor: centralize related questions model configuration by @miurla in #667
• Refactor: Improve streaming architecture for related questions generation by @miurla in #668
• feat: enhance message metadata with searchMode and modelId by @miurla in #669
• feat: emphasize Morphic reborn with agentic search in changelog by @miurla in #672
• feat: migrate model configuration to JSON files by @miurla in #673
• feat: prepare v1.0 - environment configuration cleanup by @miurla in #674
• feat: update documentation for v1.0 release by @miurla in #675
• Fix feedback button not showing after streaming completes by @miurla in #676
• fix: model selection to use cloud.json configuration properly by @miurla in #677
• feat: update prompt format and search configuration by @miurla in #678
• fix: avoid invisible action row in text parts by @miurla in #679

Full Changelog: v0.5.0-beta.3...v1.0.0-beta.4

v0.5.0-beta.3

What's Changed

• feat: update CSS variables to use oklch color space and match sidebar with muted color by @miurla in #625
• feat: align collapsible component styling for consistency by @miurla in #626
• Fix Vercel deployment error with react-markdown v9 and webpack config by @miurla in #628
• feat: add hover icon transition to SidebarTrigger and reduce header icon size by @miurla in #629
• feat: remove search mode toggle functionality by @miurla in #630
• Feature: support for Vercel gateway by @mcmadafly in #631
• feat: add icon for Vercel AI Gateway provider by @miurla in #632
• refactor: simplify provider ecosystem and model configuration by @miurla in #633
• chore: update AI SDK packages from beta to stable versions by @miurla in #634
• Improve reasoning section UI and functionality by @miurla in #635
• Update OpenAI models to GPT-5 series by @miurla in #636
• Fix text overflow in search section headers by @miurla in #637
• Fix artifact panel resize behavior to prevent chat overflow on small screens by @miurla in #638
• Redesign reasoning section UI for improved visual hierarchy by @miurla in #639
• Fix GPT-5 reasoning display configuration by @miurla in #640
• Fix OpenAI API error with GPT-5 reasoning parts in chat history by @miurla in #641
• Add category-based prompt suggestions system by @miurla in #644
• feat: add Langfuse LLM observability and tracing integration by @miurla in #645
• feat: implement Langfuse feedback integration with user rating system by @miurla in #647
• feat: add site feedback system with sentiment tracking by @miurla in #648
• feat: add changelog notification banner system by @miurla in #649
• feat: add authentication check and modal for chat access by @miurla in #650
• feat: add rate limit error modal for better user experience by @miurla in #651
• Performance optimization: Remove cache layers and add DB indexes by @miurla in #652
• perf: optimize chat API to eliminate auth duplication and improve DB operations by @miurla in #653
• Fix: Standardize AI SDK trigger and regenerate usage by @miurla in #654
• feat: add animated loading logo with eye movement by @miurla in #655

New Contributors

• @mcmadafly made their first contribution in #631

Full Changelog: v0.5.0-beta.2...v0.5.0-beta.3