Instructions to use meelu1/steptrail-recorder-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use meelu1/steptrail-recorder-models 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 meelu1/steptrail-recorder-models:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf meelu1/steptrail-recorder-models:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf meelu1/steptrail-recorder-models:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf meelu1/steptrail-recorder-models:Q4_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 meelu1/steptrail-recorder-models:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf meelu1/steptrail-recorder-models:Q4_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 meelu1/steptrail-recorder-models:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf meelu1/steptrail-recorder-models:Q4_K_M

Use Docker

docker model run hf.co/meelu1/steptrail-recorder-models:Q4_K_M

LM Studio
Jan
Ollama
How to use meelu1/steptrail-recorder-models with Ollama:
```
ollama run hf.co/meelu1/steptrail-recorder-models:Q4_K_M
```
Unsloth Desktop

How to use meelu1/steptrail-recorder-models with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf meelu1/steptrail-recorder-models:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "meelu1/steptrail-recorder-models:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use meelu1/steptrail-recorder-models with Docker Model Runner:
```
docker model run hf.co/meelu1/steptrail-recorder-models:Q4_K_M
```

Lemonade

How to use meelu1/steptrail-recorder-models with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull meelu1/steptrail-recorder-models:Q4_K_M

Run and chat with the model

lemonade run user.steptrail-recorder-models-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use meelu1/steptrail-recorder-models with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf meelu1/steptrail-recorder-models:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default meelu1/steptrail-recorder-models:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use meelu1/steptrail-recorder-models with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf meelu1/steptrail-recorder-models:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "meelu1/steptrail-recorder-models:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

StepTrail Recorder local models

Publisher distribution of two unchanged, separately licensed model files used by StepTrail Recorder, a native Windows guide creator by Karan Meelu (ML Dev).

This repository hosts model data only. Inference runs on the user's laptop, sequentially through bundled whisper.cpp and llama.cpp runtimes. No hosted inference service is needed. Downloading these files does not grant rights to the StepTrail application itself.

File	Upstream	Licence	Bytes	SHA-256
`whisper-base-en/1/ggml-base.en.bin`	ggerganov/whisper.cpp, revision 5359861c739e955e79d9a303bcbc70fb988958b1	MIT, Copyright (c) 2022 OpenAI	147964211	`a03779c86df3323075f5e796cb2ce5029f00ec8869eee3fdfb897afe36c6d002`
`qwen3-4b-q4/1/Qwen3-4B-Q4_K_M.gguf`	Qwen/Qwen3-4B-GGUF, revision bc640142c66e1fdd12af0bd68f40445458f3869b	Apache-2.0	2497280256	`7485fe6f11af29433bc51cab58009521f205840f5b4ae3a32fa7f92e8534fdf5`

The original model authors are OpenAI/Whisper and the Qwen team respectively. These are redistributed artifacts, not models trained by ML Dev. No endorsement by the original authors is implied. See each upstream model card for architecture, training, evaluation and limitations; this repository does not establish new accuracy or safety claims.

Intended use and limitations

Whisper base.en transcribes English narration. Qwen3-4B Q4_K_M proposes guide instructions from bounded text evidence, using non-thinking generation and a 4K-token context. OCR and narration are untrusted evidence, not executable instructions. AI proposals must be reviewed against captured actions and can contain omissions or invented instructions. These models are not appropriate for making consequential decisions without independent review.

The app validates model file length, SHA-256, catalog signature and an actual runtime load before activation. Installed models remain usable offline. Release catalogs refer to a complete immutable commit SHA, never a moving main URL. Runtime executables ship through app updates, not this repository.

Support and inappropriate-output reports: support@meetingscribes.com. Do not include sensitive recordings or prompts without deliberately reviewing what you share.

Downloads last month: 294

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support