Instructions to use abenzerps/K2-Horizon-3.7B-MLX-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use abenzerps/K2-Horizon-3.7B-MLX-8bit with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("abenzerps/K2-Horizon-3.7B-MLX-8bit")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use abenzerps/K2-Horizon-3.7B-MLX-8bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "abenzerps/K2-Horizon-3.7B-MLX-8bit"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "abenzerps/K2-Horizon-3.7B-MLX-8bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use abenzerps/K2-Horizon-3.7B-MLX-8bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "abenzerps/K2-Horizon-3.7B-MLX-8bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "abenzerps/K2-Horizon-3.7B-MLX-8bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "abenzerps/K2-Horizon-3.7B-MLX-8bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use abenzerps/K2-Horizon-3.7B-MLX-8bit with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "abenzerps/K2-Horizon-3.7B-MLX-8bit"

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 abenzerps/K2-Horizon-3.7B-MLX-8bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use abenzerps/K2-Horizon-3.7B-MLX-8bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "abenzerps/K2-Horizon-3.7B-MLX-8bit"

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 "abenzerps/K2-Horizon-3.7B-MLX-8bit" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

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

K2-Horizon-3.7B-MLX-8bit

File size: 1,554 Bytes

b913eda

---
base_model: IFM/K2-Horizon-3.7B
base_model_relation: quantized
license: apache-2.0
language:
  - en
pipeline_tag: text-generation
library_name: mlx
tags:
  - mlx
  - mlx-lm
  - 8-bit
  - k2-horizon
  - long-context
  - 512k-context
  - dense
---

# K2-Horizon-3.7B MLX — 8-bit

MLX 8-bit conversion of [IFM/K2-Horizon-3.7B](https://huggingface.co/IFM/K2-Horizon-3.7B), a 3.7B dense decoder-only model for reasoning, coding, long-context work, and tool use. The source checkpoint supports a native context length of **524,288 tokens (512K)**.

## Benchmarks

![IFM/K2-Horizon-3.7B benchmark results](assets/k2-horizon-3.7b-benchmarks.png)

*Benchmark results reported by IFM for the original IFM/K2-Horizon-3.7B checkpoint.*

## Release

| Format | Quantization | Size |
| --- | --- | ---: |
| MLX safetensors | Affine 8-bit, group size 64 | 5.40 GB |

The included `k2_horizon_mlx.py` adapter preserves K2 Horizon's grouped RMSNorm. Use it with MLX-LM and `--trust-remote-code`. The model is text-only; no vision projector or MTP files are included.

## Usage

```bash
pip install -U mlx-lm
mlx_lm.generate \
  --model . \
  --trust-remote-code \
  --prompt "Explain why reproducible builds matter." \
  --max-tokens 512
```

## Source

- Model: [IFM/K2-Horizon-3.7B](https://huggingface.co/IFM/K2-Horizon-3.7B)
- Source revision: [`633f52a`](https://huggingface.co/IFM/K2-Horizon-3.7B/tree/633f52ad28b17edeabd82afc61d2d13b4c59a561)
- License: [Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)
- Checksums: [SHA256SUMS.txt](SHA256SUMS.txt)