[PR #339] [CLOSED] feat: Update Anthropic model with Claude 3.7 Sonnet enhancements #1298

Closed
opened 2026-02-16 10:22:34 -05:00 by yindo · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/langgenius/dify-official-plugins/pull/339
Author: @fdb02983rhy
Created: 2/28/2025
Status: Closed

Base: mainHead: main


📝 Commits (8)

  • 6ad3c70 feat: Update Anthropic model with Claude 3.7 Sonnet enhancements
  • 6350ca5 Merge branch 'langgenius:main' into main
  • 29a9310 feat(vertex_ai): Enhance JSON schema and response handling
  • 8f91b13 Revert "feat(vertex_ai): Enhance JSON schema and response handling"
  • 3bc8d25 Update llm.py
  • 2a57d5a feat(anthropic): Enhance Claude 3.7 Sonnet model with advanced thinking block handling
  • fa9688f Merge branch 'langgenius:main' into main
  • 88fba97 Update claude-3-7-sonnet-20250219.yaml

📊 Changes

4 files changed (+272 additions, -66 deletions)

View changed files

📝 models/anthropic/manifest.yaml (+1 -1)
📝 models/anthropic/models/llm/claude-3-7-sonnet-20250219.yaml (+22 -21)
📝 models/anthropic/models/llm/llm.py (+248 -43)
📝 models/anthropic/requirements.txt (+1 -1)

📄 Description

Summary

Fix https://github.com/langgenius/dify-official-plugins/issues/318

  • Upgrade Anthropic library to version 0.48.0
  • Enhance Claude 3.7 Sonnet model configuration
    • Add new "thinking" mode parameter
    • Introduce "thinking_budget" parameter
    • Add "extended_output" beta feature support
    • Increase max tokens to 128,000
  • Improve token counting and streaming response handling
  • Bump plugin version to 0.0.8

1. Normal conversation

image

2. Tool use

image

3. PDF

image

4. Vision

image

4. Thinking mode with extended output 128k and tool use

image

5. Structured output

image

🔄 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/langgenius/dify-official-plugins/pull/339 **Author:** [@fdb02983rhy](https://github.com/fdb02983rhy) **Created:** 2/28/2025 **Status:** ❌ Closed **Base:** `main` ← **Head:** `main` --- ### 📝 Commits (8) - [`6ad3c70`](https://github.com/langgenius/dify-official-plugins/commit/6ad3c70aa53861f2e0e2a50d67240671a5343aa1) feat: Update Anthropic model with Claude 3.7 Sonnet enhancements - [`6350ca5`](https://github.com/langgenius/dify-official-plugins/commit/6350ca5f266fb9b8de53f87e286bdc1384006ef3) Merge branch 'langgenius:main' into main - [`29a9310`](https://github.com/langgenius/dify-official-plugins/commit/29a9310d82a725fc79ffdc97ab5ff78ca6126b5d) feat(vertex_ai): Enhance JSON schema and response handling - [`8f91b13`](https://github.com/langgenius/dify-official-plugins/commit/8f91b13c794ef4570fb8da85059a47ab1e3e97ae) Revert "feat(vertex_ai): Enhance JSON schema and response handling" - [`3bc8d25`](https://github.com/langgenius/dify-official-plugins/commit/3bc8d2524c6333d6f9a0e2ac9d0b622b5b133d76) Update llm.py - [`2a57d5a`](https://github.com/langgenius/dify-official-plugins/commit/2a57d5af2c70e920ec807cf34270dfd573b472a7) feat(anthropic): Enhance Claude 3.7 Sonnet model with advanced thinking block handling - [`fa9688f`](https://github.com/langgenius/dify-official-plugins/commit/fa9688f6c10f58f18ea956df4aada8e0adfdccde) Merge branch 'langgenius:main' into main - [`88fba97`](https://github.com/langgenius/dify-official-plugins/commit/88fba97738341cbb9599ae8177b563b2e9f45fb3) Update claude-3-7-sonnet-20250219.yaml ### 📊 Changes **4 files changed** (+272 additions, -66 deletions) <details> <summary>View changed files</summary> 📝 `models/anthropic/manifest.yaml` (+1 -1) 📝 `models/anthropic/models/llm/claude-3-7-sonnet-20250219.yaml` (+22 -21) 📝 `models/anthropic/models/llm/llm.py` (+248 -43) 📝 `models/anthropic/requirements.txt` (+1 -1) </details> ### 📄 Description # Summary Fix https://github.com/langgenius/dify-official-plugins/issues/318 - Upgrade Anthropic library to version 0.48.0 - Enhance Claude 3.7 Sonnet model configuration * Add new "thinking" mode parameter * Introduce "thinking_budget" parameter * Add "extended_output" beta feature support * Increase max tokens to 128,000 - Improve token counting and streaming response handling - Bump plugin version to 0.0.8 ## 1. Normal conversation <img width="747" alt="image" src="https://github.com/user-attachments/assets/ce34886a-ed68-441b-99aa-25ba62ce63ea" /> ## 2. Tool use <img width="735" alt="image" src="https://github.com/user-attachments/assets/378cf495-51f5-49cd-bcff-4529b74e3133" /> ## 3. PDF <img width="749" alt="image" src="https://github.com/user-attachments/assets/200d23ee-818f-48d2-9899-3d9632450c8a" /> ## 4. Vision <img width="613" alt="image" src="https://github.com/user-attachments/assets/9172fafd-4e55-457d-8c12-9752574cf55b" /> ## 4. Thinking mode with extended output 128k and tool use <img width="658" alt="image" src="https://github.com/user-attachments/assets/373aaee2-5227-4d66-9283-d3637464099e" /> ## 5. Structured output <img width="746" alt="image" src="https://github.com/user-attachments/assets/ab782295-ef09-4ee4-8c55-7725e2cb72e3" /> --- <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 2026-02-16 10:22:34 -05:00
yindo closed this issue 2026-02-16 10:22:34 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify-official-plugins#1298