Instructions to use datadab/writeamp-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 datadab/writeamp-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 datadab/writeamp-models:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf datadab/writeamp-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 datadab/writeamp-models:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf datadab/writeamp-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 datadab/writeamp-models:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf datadab/writeamp-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 datadab/writeamp-models:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf datadab/writeamp-models:Q4_K_M

Use Docker

docker model run hf.co/datadab/writeamp-models:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use datadab/writeamp-models with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "datadab/writeamp-models"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "datadab/writeamp-models",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/datadab/writeamp-models:Q4_K_M

Ollama
How to use datadab/writeamp-models with Ollama:
```
ollama run hf.co/datadab/writeamp-models:Q4_K_M
```
Unsloth Desktop

How to use datadab/writeamp-models with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf datadab/writeamp-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": "datadab/writeamp-models:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use datadab/writeamp-models with Docker Model Runner:
```
docker model run hf.co/datadab/writeamp-models:Q4_K_M
```

Lemonade

How to use datadab/writeamp-models with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull datadab/writeamp-models:Q4_K_M

Run and chat with the model

lemonade run user.writeamp-models-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use datadab/writeamp-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 datadab/writeamp-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 datadab/writeamp-models:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use datadab/writeamp-models with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf datadab/writeamp-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 "datadab/writeamp-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"

amitashwini commited on Jul 15

Commit

ad4d1dc

verified ·

1 Parent(s): 22d81c7

Add mini/midi Q4_K_M mirrors (mradermacher sources, byte-identical, ADR-245)

Browse files

Files changed (5) hide show

.gitattributes +2 -0
midi/README.md +45 -0
midi/SmolLM2-360M.Q4_K_M.gguf +3 -0
mini/README.md +45 -0
mini/SmolLM2-135M.Q4_K_M.gguf +3 -0

.gitattributes CHANGED Viewed

@@ -34,3 +34,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 WriteAmp-Qwen3-0.6B-Hinglish-v3.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text

 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 WriteAmp-Qwen3-0.6B-Hinglish-v3.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+midi/SmolLM2-360M.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+mini/SmolLM2-135M.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text

midi/README.md ADDED Viewed

	@@ -0,0 +1,45 @@

+---
+license: apache-2.0
+tags:
+  - text-generation
+  - smollm2
+  - gguf
+  - mirror
+base_model: HuggingFaceTB/SmolLM2-360M
+---
+# SmolLM2-360M (mirror, Q4_K_M)
+Mirror of [`mradermacher/SmolLM2-360M-GGUF`](https://huggingface.co/mradermacher/SmolLM2-360M-GGUF)'s
+`SmolLM2-360M.Q4_K_M.gguf`, hosted in WriteAmp's
+[`datadab/writeamp-models`](https://huggingface.co/datadab/writeamp-models) repo as a cold-start
+fallback for the `midi` catalog tier.
+> **Primary source of truth:** mradermacher's repo (linked above). We do **not** claim authorship of
+> the Q4_K_M quantization; this mirror is byte-identical and exists only so the `midi` tier keeps
+> working if mradermacher's URL becomes unreachable. The catalog's primary `downloadURL` continues
+> to point at mradermacher; this mirror is reached only after both `huggingface.co` and the
+> `hf-mirror.com` retry path fail (see ADR-245).
+| Field | Value |
+| --- | --- |
+| Filename | `SmolLM2-360M.Q4_K_M.gguf` |
+| Size | 270,590,464 bytes |
+| SHA-256 | `4ac7aa712c43fe7d07ed910298b5eaab932292692954abd8e0703307724e4e04` |
+| Mirror URL | `https://huggingface.co/datadab/writeamp-models/resolve/main/SmolLM2-360M.Q4_K_M.gguf?download=true` |
+## Provenance
+- **Base model:** [`HuggingFaceTB/SmolLM2-360M`](https://huggingface.co/HuggingFaceTB/SmolLM2-360M) (Apache 2.0, © Hugging Face).
+- **Q4_K_M quantization:** [`mradermacher/SmolLM2-360M-GGUF`](https://huggingface.co/mradermacher/SmolLM2-360M-GGUF). All credit for the GGUF goes to mradermacher; this mirror is a byte-identical copy made for reliability.
+## License
+Apache License 2.0. The mirror inherits the license of both the base model and mradermacher's
+Q4_K_M quantization (both Apache 2.0). Full license text lives at the base model repo.
+## Refresh policy
+This mirror is refreshed manually whenever mradermacher bumps a SHA-256. The catalog's
+`expectedSizeBytes` and `sha256` are the source of truth; an out-of-sync mirror fails the same
+size/SHA gate that already gates fresh downloads.

midi/SmolLM2-360M.Q4_K_M.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:4ac7aa712c43fe7d07ed910298b5eaab932292692954abd8e0703307724e4e04
+size 270590464

mini/README.md ADDED Viewed

	@@ -0,0 +1,45 @@

+---
+license: apache-2.0
+tags:
+  - text-generation
+  - smollm2
+  - gguf
+  - mirror
+base_model: HuggingFaceTB/SmolLM2-135M
+---
+# SmolLM2-135M (mirror, Q4_K_M)
+Mirror of [`mradermacher/SmolLM2-135M-GGUF`](https://huggingface.co/mradermacher/SmolLM2-135M-GGUF)'s
+`SmolLM2-135M.Q4_K_M.gguf`, hosted in WriteAmp's
+[`datadab/writeamp-models`](https://huggingface.co/datadab/writeamp-models) repo as a cold-start
+fallback for the `mini` catalog tier.
+> **Primary source of truth:** mradermacher's repo (linked above). We do **not** claim authorship of
+> the Q4_K_M quantization; this mirror is byte-identical and exists only so the `mini` tier keeps
+> working if mradermacher's URL becomes unreachable. The catalog's primary `downloadURL` continues
+> to point at mradermacher; this mirror is reached only after both `huggingface.co` and the
+> `hf-mirror.com` retry path fail (see ADR-245).
+| Field | Value |
+| --- | --- |
+| Filename | `SmolLM2-135M.Q4_K_M.gguf` |
+| Size | 105,454,016 bytes |
+| SHA-256 | `127c52bae983b2d0de09c60d6292743e3239e44239bbe8ccb8cf3241513ff5a0` |
+| Mirror URL | `https://huggingface.co/datadab/writeamp-models/resolve/main/SmolLM2-135M.Q4_K_M.gguf?download=true` |
+## Provenance
+- **Base model:** [`HuggingFaceTB/SmolLM2-135M`](https://huggingface.co/HuggingFaceTB/SmolLM2-135M) (Apache 2.0, © Hugging Face).
+- **Q4_K_M quantization:** [`mradermacher/SmolLM2-135M-GGUF`](https://huggingface.co/mradermacher/SmolLM2-135M-GGUF). All credit for the GGUF goes to mradermacher; this mirror is a byte-identical copy made for reliability.
+## License
+Apache License 2.0. The mirror inherits the license of both the base model and mradermacher's
+Q4_K_M quantization (both Apache 2.0). Full license text lives at the base model repo.
+## Refresh policy
+This mirror is refreshed manually whenever mradermacher bumps a SHA-256. The catalog's
+`expectedSizeBytes` and `sha256` are the source of truth; an out-of-sync mirror fails the same
+size/SHA gate that already gates fresh downloads.

mini/SmolLM2-135M.Q4_K_M.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:127c52bae983b2d0de09c60d6292743e3239e44239bbe8ccb8cf3241513ff5a0
+size 105454016