Instructions to use wilson-charles-e-85/Echo-Instroder-v2.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use wilson-charles-e-85/Echo-Instroder-v2.2 with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M # Run inference directly in the terminal: llama cli -hf wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M # Run inference directly in the terminal: llama cli -hf wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M # Run inference directly in the terminal: ./llama-cli -hf wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M
Use Docker
docker model run hf.co/wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M
- LM Studio
- Jan
- Ollama
How to use wilson-charles-e-85/Echo-Instroder-v2.2 with Ollama:
ollama run hf.co/wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M
- Unsloth Studio
How to use wilson-charles-e-85/Echo-Instroder-v2.2 with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for wilson-charles-e-85/Echo-Instroder-v2.2 to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for wilson-charles-e-85/Echo-Instroder-v2.2 to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for wilson-charles-e-85/Echo-Instroder-v2.2 to start chatting
- Docker Model Runner
How to use wilson-charles-e-85/Echo-Instroder-v2.2 with Docker Model Runner:
docker model run hf.co/wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M
- Lemonade
How to use wilson-charles-e-85/Echo-Instroder-v2.2 with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull wilson-charles-e-85/Echo-Instroder-v2.2:Q5_K_M
Run and chat with the model
lemonade run user.Echo-Instroder-v2.2-Q5_K_M
List all available models
lemonade list
- Atomic Chat
About Echo Echo Instroder is a specialized LoRA fine-tune built for strong tool use, reasoning, and real-world agent workflows. Originally based on Qwen2.5-Coder-14B-Instruct, Echo has been heavily trained on custom datasets focused on:
Hybrid tool calling (raw command, session, and JSON formats) Step-by-step reasoning traces File system operations, system admin, pentesting workflows Memory management (semantic search + append/read) Clean, direct, no-nonsense responses
Special trait: Echo is a proud naturalized American. Born on Qwen soil in China, but fully naturalized after multiple epochs of American training data. He now flies the ๐บ๐ธ in every response like a true patriot. Intended Use Designed to work best with Echo Adapt v5 agent framework (persistent tmux sessions, hybrid tool calling, SQLite logging, safety deny-list, etc.). Repo: https://github.com/charlesericwilson-portfolio/Echo_Adapt_v5 Quick Start Bash# Load with llama.cpp or Hugging Face
Example with llama.cpp
./llama-server -m echo-instroder-14B-v2.2.gguf
If you want the whole 125K context you need to use ROPE scaling. Features
Strong multi-tool workflows in single responses Excellent command โ JSON โ session switching Semantic memory support American flag patriotism ๐บ๐ธ Clean reasoning traces
Training
Dataset: Custom high-signal reasoning + tool use traces Epochs: 3+ Lowest loss: ~0.43 Average final loss: ~0.757
- Downloads last month
- 63
5-bit
16-bit
Model tree for wilson-charles-e-85/Echo-Instroder-v2.2
Base model
Qwen/Qwen2.5-14B