minions
minions copied to clipboard
HazyResearch
Big & Small LLMs working together
Added a module in `minions/utils/minion_autotrain.py` that automatically identifies local hardware capabilities, selects a suitable finetuning strategy, and finetunes an SLM from huggingface on minion_logs/ which should contain all conversations between...
# feat(evaluation): MCP Tool Evaluation for Financial Document Analysis ## Overview This Pull Request introduces a **dedicated evaluation framework** for Multi-Context Processing (MCP) tools, focusing on real-world financial documents. By...
This feature implements a "deep research" version of minions that generates the context for a query from a GPT-4o web search. The minions protocol of using this context for generating...
This PR implements an OCR parser for Minions with Gemma3. Currently, granite3.2-vision is used for OCR. Inspired by strong vision capability of the recent Gemma3 model and [a streamlit demo](https://x.com/akshay_pachaar/status/1901249805571252580)...
I created a react version of the streamlit Minions app. ### Application Structure - `App.tsx` – Main component managing global state and coordinating the app layout. - `ChatInterface.tsx` – Handles...
Added PR to create a train_minions.py file in utils which automatically detects the underlying hardware and selects the best model + training config (Full/LoRA). There are a number of todos...
### Completed Features There were a few changes I added, mostly in updating the current energy estimates for remote model inference as well as integrating this energy tracking with the...
Please check files in the following: 1. minions_deepreearch.ipynb 2. minions/minions_deepresearch.py 3. minions/minion_deepresearch.py 4. utils/serp_util.py 5. utils/firecrawl_util.py 6. README
Check out the following with my google document: https://docs.google.com/document/d/1V6_XcVElTmd_GEm9JvLhhQcnH0EYEN_fkh7ssUUtegs/edit?usp=sharing
# 1. Navigate into the electron app cd electron-app # 2. Build the frontend + Electron + create the .dmg npm run dist The resulting .dmg file will appear in:...
