Popular repositories Loading

1. DeepTalk DeepTalk Public

   A Streamlit-based PoC for inferencing DeepSeek-R1 models via OpenAI API with CoT context handling.

   Python 4

2. simple_rag simple_rag Public

   Very simple RAG example used in some of my articles and instruction.

   Python 2

3. simple_email_responder simple_email_responder Public

   An example of a simple LLM email responder.

   Python 1

4. exllamav2-calibration-dataset exllamav2-calibration-dataset Public

   Script for generating tokenized calibration datasets in Parquet format using a model's tokenizer and context length, intended for use with ExLlamaV2 quantization.

   Python 1

5. MultiStepTranslate MultiStepTranslate Public

   Simple two-stage LLM language translation.

   Python

6. minimal-deepseek-r1-rag minimal-deepseek-r1-rag Public

   A minimal two-stage Retrieval-Augmented Generation (RAG) pipeline using DeepSeek-R1 (via an OpenAI API-compatible endpoint) and DuckDuckGo for real-time web search.

   Python