Open
Conversation
…as "Text Extracted from Image"
zJuuu
added a commit
that referenced
this pull request
Mar 31, 2026
* feat: user accounts + shortcuts + new tos, privacy notice + cookie banner + enhance chat context and keyboard shortcuts functionality + Added temperature and top-p settings with local storage persistence in ChatContext. + Updated ChatProvider to retrieve and save temperature and top-p values. + Integrated keyboard shortcuts for various actions in MainLayout and ChatSidebar. + Enhanced chat components to support new props for messages and context files. + Added KeyboardShortcutsModal for improved user experience in managing shortcuts. * fix: merge err * feat: enhance folder management and data synchronization + tos, privacy notice * feat: keyboard shortcuts improvement * chore: add openai-gpt-oss-120b * chore: merge errors * feat: Add GPT-OSS and DeepSeek V3.1 (#7) * fix: fix OpenAI provider with reasoning content injection in streaming responses * fix: update fallback model ID and add deploy URL for Akash template * fix: model availability logic * fix: filter available models for display and set revalidation period * fix: enhance model availability checks and set revalidation period for static pages * fix: simplify model availability checks by removing proxy and chatapi conditions * fix: implement model ID mapping for API calls and enhance model availability checks * fix: uncomment cached models retrieval in getAvailableModels function * refactor: improve readability of model availability checks in getAvailableModels function * feat: add DeepSeek V3.1 model with enhanced capabilities and update fallback model ID * chore: comment out deploy URL for DeepSeek V3.1 * chore: cookies * feat: Add Hermes 4 405B * fix: uncomment deploy URL for Hermes 4 in model configuration * chore: ua2 + account creation * feat: models from db * fix: Fix DeepSeek V3.1 thinking and add models (#8) * fix: DeepSeek V3.1 missing chat_template args * feat: Add Qwen3 Next 80B A3B model * refactor: Update Qwen3 Next 80B A3B model details * chore: comment out deploy URL for Qwen3 Next 80B A3B model * chore: Improve private mode chat handling and toggling Adds handlePrivateModeToggle to ChatContext for centralized private mode toggling. Ensures private chats are only cleaned up when private mode is enabled and converts private chats to regular chats when disabling private mode with existing messages. Updates components to use the new toggle handler and prevents toggling private mode if messages exist and not in private mode. chore: custom auth0 domain support chore: Add local storage cleanup utility Introduced lib/local-storage-manager.ts to monitor and clean up localStorage when usage exceeds a 4MB threshold, prioritizing removal of private chats and pruning old chat messages. Integrated the cleanup check into ChatProvider initialization to help prevent storage quota issues and improve app reliability. * chore: Integrate storage error handling and improve local storage management * chore: lint * chore: rate limits from db + akashchat migration * chore: new akash handle * feat: migrate to akashml * chore: use static token first * chore: improve decryption * chore: use modelid from db * chore: update dependencies * chore: bump next * feat: dev user + db access through functions * chore: make envs optional (#10) for standalone usage with akashml api key * fix: gh action * chore: gh workflow remove version bump step * chore: update package dependencies and versions * fix: handle no available WebSocket endpoints gracefully --------- Co-authored-by: nick134 <nick134-bit@proton.me>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Added file attachment support for images. (PDF and more to come later)
Vision Model Integration for AkashChat using Ollama's LLaVA
This feature allows users to upload images to the chat and have them analyzed. Images are processed by Ollama's LLaVA vision model. The analysis is then appended to the user's message as context and sent to the AI model.
How It Works
Components
ImageUploadButton.tsx: A reusable component for the file upload buttonChatInput.tsx: Modified to include file upload and processingpages/api/vision.ts: API endpoint that processes images using Ollama's LLaVA modelutils/app/vision.ts: handles file uploads and conversionsNew Dependencies
axios: For making HTTP requests to the Ollama APIConfiguration
Install and set up Ollama:
ollama pull llavaIf your Ollama server is running on a different machine or port, update the API endpoint URL in
pages/api/vision.ts.Limitations