Minor prompt improvements. Bedrock client and keys should now be correctly passed to validation function

Added topic validation and llm deduplication functionality. Now checks for input token count before LLM calls.

Added model compatibility for OpenAI and Azure endpoints. Added some Bedrock models, now compatible with thinking models

Changed config defaults to enable successful app run without local model inference on first run

Added deduplication with LLM functionality. Minor package updates. Updated installation documentation.

Added examples for structured summaries and groups. Adapted functions for structured summaries. Simplified front tab GUI

Optimised prompts. Updated Gradio. Added example for zero shot topics. Added support for Granite 4 local model

No longer str capitalize summaries

Corrected blank KV_QUANT_LEVEL values. Removed erroneous extract topics output

Corrected KV quant value when not specified (set to None)

Corrected KV quantisation definition in config. Moved examples to top of screen under intro.

Generally improved inference for low vram systems, unsloth usage improvements, updated packages, switched default local model to Qwen 3 4b

Updated output logging files. Updated examples and readme. Minor prompt adjustments and package updates

Corrected reasoning setting

Revised intro and readme. Reasoning suffix setting simplified. All in one xlsx returns correct tables.

tracking model state through all_in_one_function

Corrected usage of stop strings, streaming

Added stop strings, optimised llama-cpp-python inference for streaming

Trying out inference with unsloth vs transformers

Added 'all-in-one' function. Corrected local model load when not loaded initially. Environment variables for max data rows and topics.

Added speculative decoding to transformers calls for gemma 3

It should now be possible to load the local model globally at the start to avoid repeated loading throughout the stages of topic extraction

Added framework of support for Azure models (although untested)

Corrected misplaced example file reference. Added AWS Nova models to model list

Minor corrections to transformers inference

Further optimisations to transformers inference

Enabled GPU-based local model inference with the transformers package

Added possibility of adding examples quickly to the input files

Enhanced app functionality by adding new UI elements for summary file management, put in Bedrock model toggle, and refined logging messages. Updated Dockerfile and requirements for better compatibility and added install guide to readme. Removed deprecated code and unnecessary comments.

Removed unnecessary print statements

Enhanced app functionality by adding new logging variables, refining file input options, and updating prompts for better user experience. Updated Dockerfile for improved environment setup and adjusted requirements for compatibility. Removed unnecessary print statements and added error handling in data loading functions.

Corrected full_prompt save

Minor fixes for Gemini, model calls. Updated Dockerfile for non-GPU systems

Merge branch 'dev' into gpt-oss-compat

GPT-OSS 20b should now work correctly

Added GPT-OSS 20b support. Moved to Llama cpp python chat_completion function

Added DynamoDB logging. Optimised topics lists with descriptions

Added cost code logging functionality. More customised logging output, can now save logs to DynamoDB.

Updated app to work correctly in API mode. Updated xlsx outputs. Some package updates

Allowed for save to xlsx file. Some package updates

You can now create structured summaries by group by changing a setting under 'I have my own list of topics'

Corrected system prompt usage. Upgraded Bedrock to converse API. Replaced Sonnet 3 with Sonnet 3.7. Updated prompts with response prefill and more specific system prompt for tables.

Further improved grouping implementation, improved summarisation prompts

Corrected LLM_MAX_GPU_LAYERS variable assignment

Added LLM_MAX_GPU_LAYERS to config

Updated requirements and config to renable Gemma 2 2B as the default for zero GPU space compatibility

Fixed some issues with output folder references. Minor changes to column names, summary output

Changed default local model to Gemma 1B

Code reorganisation to better use config files. Adapted code to use Gemma 3 as local model. Minor package updates