What's new in LangChain? đ
đ§¯ Standard exceptions: chat models now raise standard exception types across providers, so you can distinguish retryable errors (timeouts, rate limits, etc.) in a consistent way. Fully backward compatible.
đ§Š Agent middleware: custom token_counter in ContextEditingMiddleware, plus non-retryable exceptions now skip retries in ModelRetryMiddleware (
đĨ Fireworks document reranking: a new reranker integration (
đ Token & usage accounting: bug fixes for xAI and DeepSeek (
⥠Portability & perf: lazy transformers import, and tighter grep scope for Anthropic (
This is your harness! Thanks to everyone who made it better. đ