Pinned Loading

1. gRPC-video-streaming-server-golang gRPC-video-streaming-server-golang Public

   Video streaming using gRPC with no seeking forward or backward. The video is converted to protobuf which is served on HTTP to play on web app.

   HTML

2. network-packet-analyser network-packet-analyser Public

   A DNS packet monitor tool which monitors the DNS domains from a list and displays the frequency of visits on a laptop.

   Go 3 2

3. LLM-AI-Agents-Learning-Journey LLM-AI-Agents-Learning-Journey Public

   This repository is a continuously evolving collection of blogs, hands-on code, and research on Large Language Models (LLMs), AI Agents, tokenization techniques, AI security risks, and optimization …

   Jupyter Notebook 4

4. threat-detection-analysis-tools threat-detection-analysis-tools Public

   This repo contains tools for threat detection and analysis of various malwares or any such suspicious activities.

   Go 1 3

5. 100_models_grouped_by_vram 100_models_grouped_by_vram

   1

   |id |model_name                    |url                                                                 |type           |params_billion|typical_use                |est_vram_4bit_gb|approx_vram_gb|vram_tier |kv_cache_support                  |sharding                                                          |quant_formats                                                  |determinism_tip                                                                                                    |good_to_know                                                                                               |notes                                  |

   2

   |---|------------------------------|--------------------------------------------------------------------|---------------|--------------|---------------------------|----------------|--------------|----------|----------------------------------|------------------------------------------------------------------|---------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------|---------------------------------------|

   3

   |1  |Llama-3.1-8B-Instruct         |https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct             |LLM            |8.0           |General chat / coding      |6–7             |6.0           |4–<8 GB   |Yes (Transformer decoder; vLLM/HF)|Not needed for single-GPU                                         |GGUF (llama.cpp), AWQ, GPTQ, bitsandbytes 4/8 where available  |For repeatability: temperature 0.0–0.2, top_p=1.0, fix random seed; use greedy/beam for most deterministic outputs.|General: use 4-bit quant + paged KV cache; keep batch small on consumer GPUs.                              |128K ctx family; popular 8B.           |

   4

   |2  |Mistral-7B-v0.3               |https://huggingface.co/mistralai/Mistral-7B-v0.3                    |LLM            |7.3           |General chat               |5–6             |5.0           |4–<8 GB   |Yes (Transformer decoder; vLLM/HF)|Not needed for single-GPU                                         |GGUF (llama.cpp), AWQ, GPTQ, bitsandbytes 4/8 where available  |For repeatability: temperature 0.0–0.2, top_p=1.0, fix random seed; use greedy/beam for most deterministic outputs.|General: use 4-bit quant + paged KV cache; keep batch small on consumer GPUs.                              |Efficient 7B; many finetunes.          |

   5

   |3  |Qwen2.5-7B-Instruct           |https://huggingface.co/Qwen/Qwen2.5-7B-Instruct                     |LLM            |7.0           |General chat               |5–6             |5.0           |4–<8 GB   |Yes (Transformer decoder; vLLM/HF)|Not needed for single-GPU                                         |GGUF (llama.cpp), AWQ, GPTQ, bitsandbytes 4/8 where available  |For repeatability: temperature 0.0–0.2, top_p=1.0, fix random seed; use greedy/beam for most deterministic outputs.|General: use 4-bit quant + paged KV cache; keep batch small on consumer GPUs.                              |Long-context variants exist.           |