Skip to content

Conversation

@irony
Copy link

@irony irony commented Jan 23, 2026

Summary

Add Berget.AI as an OpenAI-compatible provider with base URL api.berget.ai/v1.

Models Added

  • Text Models: Llama 3.3 70B Instruct, Qwen3 32B, GPT-OSS-120B, GLM 4.7, Mistral Small 3.2 24B Instruct 2506
  • Embedding Models: Multilingual-E5-large-instruct, Multilingual-E5-large
  • Rerank Model: bge-reranker-v2-m3
  • Speech-to-Text Model: KB-Whisper-Large

Details

  • Provider configuration at providers/berget/provider.toml
  • Uses @ai-sdk/openai-compatible npm package
  • Environment variable: BERGET_API_KEY
  • Documentation: https://api.berget.ai

Add Berget.AI as an OpenAI-compatible provider with base URL api.berget.ai/v1.

Models included:
- Text: Llama 3.3 70B, Qwen3 32B, GPT-OSS-120B, GLM 4.7, Mistral Small 3.2 24B
- Embedding: Multilingual-E5-large-instruct, Multilingual-E5-large
- Rerank: bge-reranker-v2-m3
- Speech-to-Text: KB-Whisper-Large
Copilot AI review requested due to automatic review settings January 23, 2026 00:03
Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This pull request adds Berget.AI as a new OpenAI-compatible provider with comprehensive model support across text generation, embeddings, reranking, and speech-to-text capabilities.

Changes:

  • Adds Berget.AI provider configuration using @ai-sdk/openai-compatible npm package
  • Configures 5 text generation models (Llama 3.3 70B, Qwen3 32B, GPT-OSS-120B, GLM 4.7, Mistral Small 3.2)
  • Adds 2 embedding models, 1 reranker model, and 1 speech-to-text model
  • Includes provider logo in SVG format

Reviewed changes

Copilot reviewed 9 out of 10 changed files in this pull request and generated 3 comments.

Show a summary per file
File Description
providers/berget/provider.toml Provider configuration with API endpoint, npm package, and environment variable
providers/berget/models/zai-org/GLM-4.7.toml GLM 4.7 text generation model configuration
providers/berget/models/openai/gpt-oss-120b.toml GPT-OSS-120B text generation model configuration
providers/berget/models/mistralai/Mistral-Small-3.2-24B-Instruct-2506.toml Mistral Small 3.2 text generation model configuration
providers/berget/models/meta-llama/Llama-3.3-70B-Instruct.toml Llama 3.3 70B Instruct text generation model configuration
providers/berget/models/intfloat/multilingual-e5-large.toml Multilingual E5 large embedding model configuration
providers/berget/models/intfloat/multilingual-e5-large-instruct.toml Multilingual E5 large instruct embedding model configuration
providers/berget/models/Qwen/Qwen3-32B.toml Qwen3 32B text generation model configuration
providers/berget/models/KBLab/kb-whisper-large.toml KB-Whisper-Large speech-to-text model configuration
providers/berget/models/BAAI/bge-reranker-v2-m3.toml BGE reranker v2-m3 model configuration
providers/berget/logo.svg Berget.AI provider logo

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

- Change Mistral family from 'mistral' to 'mistral-small' for consistency
- Fix Llama 3.3 70B knowledge date from '2024-12' to '2023-12'
- Set tool_call to false for KB-Whisper-Large (speech-to-text models don't support tool calling)
@irony
Copy link
Author

irony commented Jan 26, 2026

Any news on when to merge this or any review comments?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant