[PR #332] [MERGED] Infinite prompt input and compression implementation #3301

Closed
opened 2026-02-22 18:33:30 -05:00 by yindo · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/Mintplex-Labs/anything-llm/pull/332
Author: @timothycarambat
Created: 11/3/2023
Status: Merged
Merged: 11/6/2023
Merged by: @timothycarambat

Base: masterHead: infinite-prompt-input


📝 Commits (8)

  • 77152b4 WIP on continuous prompt window summary
  • 34ca067 wip
  • 29635a1 Move chat out of VDB
  • 2ac1367 Implement compression for Anythropic
  • 9ad0c01 cleanup vectorDBs and check that lance, chroma, and pinecone are returning valid metadata sources
  • 306b112 Resolve Weaviate citation sources not working with schema
  • 808b8ee merge with master
  • 5fb7fd9 comment cleanup

📊 Changes

23 files changed (+830 additions, -438 deletions)

View changed files

📝 frontend/src/components/LLMSelection/AnthropicAiOptions/index.jsx (+2 -2)
📝 frontend/src/components/LLMSelection/AzureAiOptions/index.jsx (+17 -0)
📝 frontend/src/components/Modals/MangeWorkspace/Settings/index.jsx (+0 -1)
📝 frontend/src/components/WorkspaceChat/ChatContainer/PromptInput/index.jsx (+0 -1)
📝 server/endpoints/chat.js (+1 -0)
server/models/cacheData.js (+69 -0)
📝 server/models/systemSettings.js (+1 -0)
📝 server/package.json (+1 -0)
server/prisma/migrations/20231101195421_init/migration.sql (+11 -0)
📝 server/prisma/schema.prisma (+11 -0)
📝 server/utils/AiProviders/anthropic/index.js (+45 -18)
📝 server/utils/AiProviders/azureOpenAi/index.js (+40 -20)
📝 server/utils/AiProviders/openAi/index.js (+37 -11)
📝 server/utils/chats/index.js (+127 -71)
server/utils/helpers/chat/index.js (+325 -0)
server/utils/helpers/tiktoken.js (+57 -0)
📝 server/utils/helpers/updateENV.js (+13 -1)
📝 server/utils/vectorDbProviders/chroma/index.js (+9 -67)
📝 server/utils/vectorDbProviders/lance/index.js (+18 -62)
📝 server/utils/vectorDbProviders/pinecone/index.js (+13 -61)

...and 3 more files

📄 Description

Split LLM model fully from chat
Add compression methods to all LLMs to ensure inifinite message sizing will work
Refactor VectorDBs to only provide interfaces that have to do with vectorization
Add polymorphic cache_data table
Update Anthropic model selection
Enabled token limit selection for Azure


🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/Mintplex-Labs/anything-llm/pull/332 **Author:** [@timothycarambat](https://github.com/timothycarambat) **Created:** 11/3/2023 **Status:** ✅ Merged **Merged:** 11/6/2023 **Merged by:** [@timothycarambat](https://github.com/timothycarambat) **Base:** `master` ← **Head:** `infinite-prompt-input` --- ### 📝 Commits (8) - [`77152b4`](https://github.com/Mintplex-Labs/anything-llm/commit/77152b4d649f7387456958e560220b9ac8dc4a9d) WIP on continuous prompt window summary - [`34ca067`](https://github.com/Mintplex-Labs/anything-llm/commit/34ca06710b6c36becdd568ca1105ef5ded3f5200) wip - [`29635a1`](https://github.com/Mintplex-Labs/anything-llm/commit/29635a13e17b15495ec11161677ceea91372a890) Move chat out of VDB - [`2ac1367`](https://github.com/Mintplex-Labs/anything-llm/commit/2ac136745028a751c83304154a47db986bb1c2dd) Implement compression for Anythropic - [`9ad0c01`](https://github.com/Mintplex-Labs/anything-llm/commit/9ad0c0176090ca340d31a630a59bf9ab6c1e4fc2) cleanup vectorDBs and check that lance, chroma, and pinecone are returning valid metadata sources - [`306b112`](https://github.com/Mintplex-Labs/anything-llm/commit/306b112ba2dad63b52a2de5dc1295e0959df5226) Resolve Weaviate citation sources not working with schema - [`808b8ee`](https://github.com/Mintplex-Labs/anything-llm/commit/808b8eeb4ec3e681c30c159cef0d248726776cc1) merge with master - [`5fb7fd9`](https://github.com/Mintplex-Labs/anything-llm/commit/5fb7fd9abfd773eaff011ed5c83568b66f2b0231) comment cleanup ### 📊 Changes **23 files changed** (+830 additions, -438 deletions) <details> <summary>View changed files</summary> 📝 `frontend/src/components/LLMSelection/AnthropicAiOptions/index.jsx` (+2 -2) 📝 `frontend/src/components/LLMSelection/AzureAiOptions/index.jsx` (+17 -0) 📝 `frontend/src/components/Modals/MangeWorkspace/Settings/index.jsx` (+0 -1) 📝 `frontend/src/components/WorkspaceChat/ChatContainer/PromptInput/index.jsx` (+0 -1) 📝 `server/endpoints/chat.js` (+1 -0) ➕ `server/models/cacheData.js` (+69 -0) 📝 `server/models/systemSettings.js` (+1 -0) 📝 `server/package.json` (+1 -0) ➕ `server/prisma/migrations/20231101195421_init/migration.sql` (+11 -0) 📝 `server/prisma/schema.prisma` (+11 -0) 📝 `server/utils/AiProviders/anthropic/index.js` (+45 -18) 📝 `server/utils/AiProviders/azureOpenAi/index.js` (+40 -20) 📝 `server/utils/AiProviders/openAi/index.js` (+37 -11) 📝 `server/utils/chats/index.js` (+127 -71) ➕ `server/utils/helpers/chat/index.js` (+325 -0) ➕ `server/utils/helpers/tiktoken.js` (+57 -0) 📝 `server/utils/helpers/updateENV.js` (+13 -1) 📝 `server/utils/vectorDbProviders/chroma/index.js` (+9 -67) 📝 `server/utils/vectorDbProviders/lance/index.js` (+18 -62) 📝 `server/utils/vectorDbProviders/pinecone/index.js` (+13 -61) _...and 3 more files_ </details> ### 📄 Description Split LLM model fully from chat Add compression methods to all LLMs to ensure inifinite message sizing will work Refactor VectorDBs to only provide interfaces that have to do with vectorization Add polymorphic cache_data table Update Anthropic model selection Enabled token limit selection for Azure --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
yindo added the pull-request label 2026-02-22 18:33:30 -05:00
yindo closed this issue 2026-02-22 18:33:30 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#3301