v0.2.0

What's New

New Features

• Vision/Image Support - Analyze images with multimodal models

  • New create_with_image() convenience method
  • Support for URLs and local files (auto base64 encoding)
  • Multiple image support

• Audio I/O - Audio input and output support

  • Audio input in messages
  • Audio output with voice selection

• Video Support - Video content analysis

• Web Search - Real-time web search integration

  • New create_with_web_search() convenience method
  • Citation/annotation support

• Reasoning Mode - Chain-of-thought reasoning for complex problems (glm-4.7)

• Extended Thinking - Deep thinking for Gemini models

• Models API - List and retrieve available models

  • client.models.list()
  • client.models.retrieve(model_id)

• Responses API - OpenAI Responses API format

  • client.responses.create()

• Messages API - Anthropic-native message format

  • client.messages.create()

• Rerank API - Document reranking for RAG applications

  • client.rerank.create()
  • Supports BAAI/bge-reranker-v2-m3, Qwen/Qwen3-Reranker models

• Helper Utilities - Base64 encoding for local files

Testing

• 81 unit tests covering all new functionality

Installation

pip install apertis==0.2.0

Full Changelog: v0.1.1...v0.2.0