Instructions to use lthn/lemer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lthn/lemer with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="lthn/lemer",
	filename="lemer-bf16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": [
				{
					"type": "text",
					"text": "Describe this image in one sentence."
				},
				{
					"type": "image_url",
					"image_url": {
						"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
					}
				}
			]
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use lthn/lemer with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf lthn/lemer:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf lthn/lemer:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf lthn/lemer:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf lthn/lemer: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 lthn/lemer:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf lthn/lemer: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 lthn/lemer:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf lthn/lemer:Q4_K_M

Use Docker

docker model run hf.co/lthn/lemer:Q4_K_M

LM Studio
Jan

vLLM

How to use lthn/lemer with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lthn/lemer"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lthn/lemer",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/lthn/lemer:Q4_K_M

Ollama
How to use lthn/lemer with Ollama:
```
ollama run hf.co/lthn/lemer:Q4_K_M
```

Unsloth Studio new

How to use lthn/lemer 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 lthn/lemer 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 lthn/lemer to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for lthn/lemer to start chatting

Pi new

How to use lthn/lemer with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf lthn/lemer:Q4_K_M

Configure the model in Pi

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

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use lthn/lemer with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf lthn/lemer: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 lthn/lemer:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use lthn/lemer with Docker Model Runner:
```
docker model run hf.co/lthn/lemer:Q4_K_M
```

Lemonade

How to use lthn/lemer with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull lthn/lemer:Q4_K_M

Run and chat with the model

lemonade run user.lemer-Q4_K_M

List all available models

lemonade list

Snider Cladius Maximus commited on Apr 9

Commit

f526be0

1 Parent(s): dd31d86

feat: add Ollama config files + complete repo file documentation

Browse files

- template: explicit Go template for Ollama chat formatting
- params: sampling parameters (temp=1.0, top_p=0.95, top_k=64, stop tokens)
- README: full repo file table, MLX quick start, conversational tag
- Tags: added safetensors, mlx, conversational for HF discoverability

Explicit config > auto-detect for every supported ecosystem.

Co-Authored-By: Cladius Maximus <cladius@lethean.io>

Files changed (3) hide show

README.md +28 -1
params +6 -0
template +7 -0

README.md CHANGED Viewed

@@ -12,11 +12,14 @@ tags:
 - gemma4
 - gemma
 - gguf
 - lemma
 - lethean
 - lem
 - apple-silicon
 - on-device
 datasets:
 - lthn/LEM-research
 - lthn/livebench
@@ -41,6 +44,23 @@ The smallest member of the [Lemma model family](https://huggingface.co/collectio
 All variants tested and verified working with Ollama on Apple Silicon (M3 Ultra, 96GB).
 ## Quick Start
 ### Ollama
@@ -52,7 +72,14 @@ ollama run hf.co/lthn/lemer:Q4_K_M
 ### llama.cpp
 ```bash
-llama-cli -m lemer-q4_k_m.gguf -p "Hello, how are you?" -n 200
 ```
 ## Model Details

 - gemma4
 - gemma
 - gguf
+- safetensors
+- mlx
 - lemma
 - lethean
 - lem
 - apple-silicon
 - on-device
+- conversational
 datasets:
 - lthn/LEM-research
 - lthn/livebench
 All variants tested and verified working with Ollama on Apple Silicon (M3 Ultra, 96GB).
+This repo also includes MLX Q4 safetensors for native Apple Silicon inference via `mlx-lm`. See [lemer-mlx-q8](https://huggingface.co/lthn/lemer-mlx-q8) and [lemer-mlx-bf16](https://huggingface.co/lthn/lemer-mlx-bf16) for other MLX quant levels.
+### Repo Files
+| File | Format | Purpose |
+|------|--------|---------|
+| `lemer-*.gguf` | GGUF | Ollama, llama.cpp, GPT4All, LM Studio |
+| `model.safetensors` | MLX safetensors | Native Apple Silicon via `mlx-lm` (Q4) |
+| `config.json` | JSON | Model architecture config |
+| `tokenizer.json` | JSON | Tokenizer vocabulary |
+| `tokenizer_config.json` | JSON | Tokenizer settings |
+| `chat_template.jinja` | Jinja2 | Chat template for transformers/mlx-lm |
+| `processor_config.json` | JSON | Image/audio processor config |
+| `generation_config.json` | JSON | Default generation parameters |
+| `template` | Go template | Ollama chat template override |
+| `params` | JSON | Ollama sampling parameters |
 ## Quick Start
 ### Ollama
 ### llama.cpp
 ```bash
+llama-cli -hf lthn/lemer:Q4_K_M
+```
+### MLX (Apple Silicon native)
+```bash
+pip install mlx-lm
+mlx_lm.generate --model lthn/lemer --prompt "Hello, how are you?" --max-tokens 200
 ```
 ## Model Details

params ADDED Viewed

	@@ -0,0 +1,6 @@

+{
+    "temperature": 1.0,
+    "top_p": 0.95,
+    "top_k": 64,
+    "stop": ["<turn|>", "<eos>"]
+}

template ADDED Viewed

	@@ -0,0 +1,7 @@

+{{- if .System }}<|turn>system
+{{ .System }}<turn|>
+{{ end -}}
+<|turn>user
+{{ .Prompt }}<turn|>
+<|turn>model
+{{ .Response }}