LlamaIndex v0.14.5
LlamaIndex v0.14.5 releases with bug fixes, progress bar support for multiprocess loading, and expanded model support including GPT-5, Claude Haiku 4.5, and Anthropic prompt caching utilities.
Release Notes [2025-10-15] llama-index-core [0.14.5] Remove debug print (#20000) safely initialize RefDocInfo in Docstore (#20031) Add progress bar for multiprocess loading (#20048) Fix duplicate node positions when identical text appears multiple times in document (#20050) chore: tool call block - part 1 (#20074) llama-index-instrumentation [0.4.2] update instrumentation package metadata (#20079) llama-index-llms-anthropic [0.9.5] ✨ feat(anthropic): add prompt caching model validation utilities (#20069) fix streaming thinking/tool calling with anthropic (#20077) Add haiku 4.5 support (#20092) llama-index-llms-baseten [0.1.6] Baseten provider Kimi K2 0711, Llama 4 Maverick and Llama 4 Scout Model APIs deprecation (#20042) llama-index-llms-bedrock-converse [0.10.5] feat: List Claude Sonnet 4.5 as a reasoning model (#20022) feat: Support global cross-region inference profile prefix (#20064) Update utils.py for opus 4.1 (#20076) 4.1 opus bedrockconverse missing in funcitoncalling models (#20084) Add haiku 4.5 support (#20092) llama-index-llms-fireworks [0.4.4] Add Support for Custom Models in Fireworks LLM (#20023) fix(llms/fireworks): Cannot use Fireworks Deepseek V3.1-20006 issue…
- github.comllama_index v0.14.5primary