the-crypt-keeper

can-ai-code. Self-evaluating interview for AI coders

tldw. Too Long, Didn't Watch: End-to-End Rolling Summarizer of Long Videos

LLooM. Experimental LLM Inference UX to aid in creative writing

reasonscape. Information Processing Evaluation for Large Language Models

tiny_starcoder. Python examples using the bigcode/tiny_starcoder_py 159M model to generate code

the-muse. Experimental sampler to make LLMs more creative

ggml-downloader. Simple, Fast, Parallel Huggingface GGML model downloader written in python

tcurtsni. Tcurtsni: Reverse Instruction Chat, ever wonder what your LLM wants to ask you?

ChatBench. Evaluating practical performance of local multi-turn conversational LLMs.

llm-webapps. jQuery, React and Streamlit applications written by LLMs

fragment-frog. 🐸 A non-linear AI-enabled text editing environment

llama-srb-api. Single Request Batching API server backed by llama.cpp

modelzoo. ModelZoo is a system for managing and serving local Large Language Models (LLMs).

llm-visualizer. Interactive LLM weight explorer made with Streamlit

llproxy. Automated proxy to merge multiple LLM endpoints into one

rosettacode-parser. Scripts to parse rosettacode scrape into structured data

Medusa-Bridge. Join multiple local LLM API servers to the AI Horde

LlamaCpp-Horde-Bridge. Use llama.cpp server to join the AI Horde

bluey-llm. Training a micro transformers model from scratch to write "Bluey" episodes

NBCE. Naive Bayes-based Context Extension

dell-r730-ipmi-hacks. Scripts to make the Dell R730 behave properly with Nvidia Pascal P40 and other datacenter GPUs

cascade. Async Generative Workflows Manager

scribe. Simple Synthetic Data Generation with LLMs

battlerealms. Exploring BBS-style strategy gaming as a bounded challenge.

ggml. Tensor library for machine learning

toe2toe. Large Language Models playing Tic-Tac-Toe against each other

broken-record. Synthetic LLM Conversation Analyzer

param-optimizer. Evolution/Genetic Algorithms applied to finding ideal LLM Sampling Parameters