Instructions to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit 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("SirSahOl/NeoHorse-1-9B-chat-mlx-4bit")

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)

Transformers

How to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="SirSahOl/NeoHorse-1-9B-chat-mlx-4bit")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SirSahOl/NeoHorse-1-9B-chat-mlx-4bit")
model = AutoModelForCausalLM.from_pretrained("SirSahOl/NeoHorse-1-9B-chat-mlx-4bit", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

vLLM

How to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SirSahOl/NeoHorse-1-9B-chat-mlx-4bit

SGLang

How to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

How to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit"

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": "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Docker Model Runner
How to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit with Docker Model Runner:
```
docker model run hf.co/SirSahOl/NeoHorse-1-9B-chat-mlx-4bit
```

Hermes Agent

How to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit 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 "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit"

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 SirSahOl/NeoHorse-1-9B-chat-mlx-4bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use SirSahOl/NeoHorse-1-9B-chat-mlx-4bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit"

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 "SirSahOl/NeoHorse-1-9B-chat-mlx-4bit" \
  --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"

NeoHorse-1-9B-mlx-4bit

4-bit MLX conversion of TokenRhythm/NeoHorse-1-9B optimized for Apple Silicon native GPU inference.

Converted by: SirSahOl
Source Model: TokenRhythm/NeoHorse-1-9B
Framework: MLX by Apple
Quantization: 4-bit (Average 4.50 bits per weight)
Format: safetensors
License: apache-2.0

Model Details

Architecture: Qwen3_5ForCausalLM
Parameters: 9.0B
Context Length: 262,144 tokens
Format: MLX (Apple Silicon native GPU format)
Quantization: 4-bit (Average 4.50 bits per weight)
Active VRAM Footprint: ~5.4 GB (Minimum recommended: 16 GB Unified Memory)

Quick Start

Installation

pip install mlx-lm

CLI Usage

# Chat interactively
mlx_lm.chat --model SirSahOl/NeoHorse-1-9B-chat-mlx-4bit

# Generate text
mlx_lm.generate --model SirSahOl/NeoHorse-1-9B-chat-mlx-4bit --prompt "Write a short poem about artificial intelligence."

Python API (with Chat Template)

from mlx_lm import load, generate

model, tokenizer = load("SirSahOl/NeoHorse-1-9B-chat-mlx-4bit")

messages = [
    {"role": "user", "content": "Explain quantum superposition in simple terms."}
]
prompt = tokenizer.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True
)

response = generate(model, tokenizer, prompt=prompt, verbose=True)
print(response)

Performance Benchmarks

Apple Silicon Hardware Sizing Matrix

Estimated decoding throughput, time-to-first-token (TTFT), and active unified memory footprint across Apple Silicon tiers:

Apple Silicon Tier	Unified Memory	Active VRAM	Estimated Speed	Est. TTFT	Recommended Use Case
M1 / M2 / M3 / M4 (Base)	16 GB Unified Memory	~5.4 GB	~28 tokens/sec	~141 ms	Everyday interactive assistant & fast local completions
M1 / M2 / M3 / M4 Pro	18 GB – 36 GB	~5.4 GB	~42 tokens/sec	~96 ms	Balanced daily driver for coding, tool invocation, and multi-turn chat
M1 / M2 / M3 / M4 Max	36 GB – 128 GB	~5.4 GB	~60 tokens/sec	~59 ms	High-throughput generation, low latency, agent orchestration
M1 / M2 / M3 Ultra	64 GB – 192 GB	~5.4 GB	~84 tokens/sec	~39 ms	Peak concurrency, batch document extraction, production serving

Estimates based on Apple Silicon unified memory bandwidth and active parameter footprint. Real-world speeds may vary with context length.

Multi-Quantization Comparison

Evaluate your hardware budget and choose the optimal precision:

Variant	Disk Size	VRAM Footprint	Target Apple Silicon Hardware	Key Advantage
4-bit MLX (This Repository)	~5.3 GB	~5.3 GB	M1 / M2 / M3 / M4 (8GB+)	Maximum generation speed and lowest RAM overhead.
8-bit MLX	~9.9 GB	~9.9 GB	M1 / M2 / M3 / M4 Pro/Max (16GB+)	Balanced accuracy and generation speed; near-lossless reasoning.
16-bit MLX	~18.8 GB	~18.8 GB	M2 / M3 / M4 Max/Ultra (32GB+)	Full unquantized precision; reference evaluation quality.

Who Should Use This?

Your Hardware	Recommended Quantization
M1/M2/M3/M4 (8GB – 16GB)	4-bit — Best balance of speed, low memory, and multitasking capability
M1/M2/M3/M4 Pro/Max (18GB – 36GB)	8-bit — Higher quality reasoning with comfortable memory headroom
M1/M2/M3/M4 Max/Ultra (36GB – 192GB)	16-bit — Unquantized full precision, zero quality degradation

General guidance:

Use 4-bit if you want to run this model alongside IDEs, browsers, and background development tools.
Use 8-bit if you have 16GB+ unified memory and require superior reasoning and code accuracy.
Use 16-bit for research, benchmarking, evaluation, or high-end workstation deployments.

Other Quantization Variants

Variant	Link
4-bit	SirSahOl/NeoHorse-1-9B-chat-mlx-4bit
8-bit	SirSahOl/NeoHorse-1-9B-chat-mlx-8bit
16-bit	SirSahOl/NeoHorse-1-9B-chat-mlx-16bit

LM Studio & Local Inference Setup Guide

To prevent runaway loops and ensure correct conversational turn-taking, configure custom stop strings in your local inference runtime.

Custom Stop Strings Configuration

Ensure the following sequence tokens are configured as strict stop boundaries:

<|im_start|>
<|im_end|>
<|endoftext|>

Option A: Automatic Preset (Recommended for LM Studio)

Create a custom prompt preset JSON file named ChatML.json in your LM Studio config directory:

macOS / Linux: ~/.lmstudio/config-presets/ChatML.json
Windows: %USERPROFILE%\.lmstudio\config-presets\ChatML.json

{
  "name": "ChatML",
  "inference_params": {
    "pre_prompt": "You are a helpful, respectful, and honest assistant.",
    "input_prefix": "<|im_start|>user\n",
    "input_suffix": "<|im_end|>\n",
    "pre_prompt_prefix": "<|im_start|>system\n",
    "pre_prompt_suffix": "<|im_end|>\n",
    "antiprompt": [
      "<|im_start|>",
      "<|im_end|>",
      "<|endoftext|>"
    ],
    "stopStrings": [
      "<|im_start|>",
      "<|im_end|>",
      "<|endoftext|>"
    ],
    "temperature": 0.7,
    "max_tokens": 2048
  }
}

Restart LM Studio, load SirSahOl/NeoHorse-1-9B-chat-mlx-4bit, and select "ChatML" from the Prompt Template dropdown.

Option B: Manual LM Studio Configuration

Load SirSahOl/NeoHorse-1-9B-chat-mlx-4bit in LM Studio.
In the right-hand Advanced Configuration / Inference Parameters panel:
- Stop Strings (antiprompt / stopStrings): Add <|im_start|>, <|im_end|>, <|endoftext|>
- System Prefix: <|im_start|>system\n
- System Suffix: <|im_end|>\n
- User Prefix: <|im_start|>user\n
- Assistant Suffix: <|im_end|>\n<|im_start|>assistant\n

Option C: Ollama Modelfile Setup

Create a Modelfile to run this model in Ollama:

FROM SirSahOl/NeoHorse-1-9B-chat-mlx-4bit

PARAMETER stop "<|im_start|>"
PARAMETER stop "<|im_end|>"
PARAMETER stop "<|endoftext|>"
PARAMETER temperature 0.7

TEMPLATE """{{ if .System }}<|im_start|>system
{{ .System }}<|im_end|>
{{ end }}{{ if .Prompt }}<|im_start|>user
{{ .Prompt }}<|im_end|>
{{ end }}<|im_start|>assistant
"""

Create and run with Ollama:

ollama create neohorse-1-9b-chat-mlx-4bit -f Modelfile
ollama run neohorse-1-9b-chat-mlx-4bit

Conversion Details

Property	Value
Source Model	TokenRhythm/NeoHorse-1-9B
Quantization	4-bit
mlx-lm Version	0.31.3
Conversion Time	101.37s
Output Size	4.7 GB
Date	2026-09-14T23:31:53.987005+00:00

Reproduction

To reproduce this conversion:

pip install mlx-lm==0.31.3
python3 -m mlx_lm.convert --hf-path /root/.cache/huggingface/hub/models--TokenRhythm--NeoHorse-1-9B/snapshots/ba5b6e40d88a6ddf4591e176738254a3bc715765 --mlx-path output/NeoHorse-1-9B-mlx-4bit -q --q-bits 4

Limitations & Known Issues

4-bit group-wise quantization introduces minor precision loss compared to unquantized weights; for deep mathematical derivations or precision-critical reasoning, test the 8-bit or 16-bit variants.
High context sequences (>32K tokens) require sufficient unified memory headroom; ensure unified memory is not overcommitted.
This is a weight-only MLX conversion designed specifically for Apple Silicon GPUs (M1/M2/M3/M4 series).