Text Generation
MLX
Safetensors
English
Chinese
glm_moe_dsa
glm
glm-5
apple-silicon
quantized
2-8bit
Mixture of Experts
orcasaq
dynamic-quant
reasoning
coding
agentic
conversational
4-bit precision
Instructions to use orcarouter/GLM-5.3-MLX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use orcarouter/GLM-5.3-MLX 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("orcarouter/GLM-5.3-MLX") 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
- Pi
How to use orcarouter/GLM-5.3-MLX with Pi:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "orcarouter/GLM-5.3-MLX"
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": "orcarouter/GLM-5.3-MLX" } ] } } }Run Pi
# Start Pi in your project directory: pi
- MLX LM
How to use orcarouter/GLM-5.3-MLX with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Interactive chat REPL mlx_lm.chat --model "orcarouter/GLM-5.3-MLX"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "orcarouter/GLM-5.3-MLX" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "orcarouter/GLM-5.3-MLX", "messages": [ {"role": "user", "content": "Hello"} ] }' - Hermes Agent
How to use orcarouter/GLM-5.3-MLX 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 "orcarouter/GLM-5.3-MLX"
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 orcarouter/GLM-5.3-MLX
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use orcarouter/GLM-5.3-MLX with OpenClaw:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "orcarouter/GLM-5.3-MLX"
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 "orcarouter/GLM-5.3-MLX" \ --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"
File size: 2,370 Bytes
9a5177c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 | {
"attention@8bit": {
"n": 390,
"cos_mean": 0.9999843754304225,
"cos_min": 0.9999701142781648,
"cos_p5": 0.9999817950797752,
"snr_mean": 45.05945505567317,
"worst": [
"model.layers.0.self_attn.q_b_proj.weight",
"model.layers.7.self_attn.q_b_proj.weight",
"model.layers.2.self_attn.q_b_proj.weight",
"model.layers.3.self_attn.q_b_proj.weight",
"model.layers.3.self_attn.kv_a_proj_with_mqa.weight"
]
},
"dense_mlp@6bit": {
"n": 9,
"cos_mean": 0.9997481794588633,
"cos_min": 0.9997417464914627,
"cos_p5": 0.9997417464914627,
"snr_mean": 32.974673175379735,
"worst": [
"model.layers.2.mlp.down_proj.weight",
"model.layers.1.mlp.down_proj.weight",
"model.layers.0.mlp.down_proj.weight",
"model.layers.1.mlp.up_proj.weight",
"model.layers.2.mlp.gate_proj.weight"
]
},
"expert_down@5bit": {
"n": 19200,
"cos_mean": 0.9989847256501002,
"cos_min": 0.9981335686868937,
"cos_p5": 0.9989747346453054,
"snr_mean": 26.926658932773485,
"worst": [
"model.layers.4.mlp.experts.96.down_proj.weight",
"model.layers.6.mlp.experts.87.down_proj.weight",
"model.layers.3.mlp.experts.200.down_proj.weight",
"model.layers.5.mlp.experts.234.down_proj.weight",
"model.layers.8.mlp.experts.86.down_proj.weight"
]
},
"expert_gate_up@4bit": {
"n": 38400,
"cos_mean": 0.9956952468956632,
"cos_min": 0.9950926262273702,
"cos_p5": 0.9955739241692824,
"snr_mean": 20.655963162017926,
"worst": [
"model.layers.4.mlp.experts.168.gate_proj.weight",
"model.layers.3.mlp.experts.2.up_proj.weight",
"model.layers.4.mlp.experts.168.up_proj.weight",
"model.layers.3.mlp.experts.2.gate_proj.weight",
"model.layers.4.mlp.experts.186.gate_proj.weight"
]
},
"shared_expert@6bit": {
"n": 225,
"cos_mean": 0.9997443424916399,
"cos_min": 0.9996723558344117,
"cos_p5": 0.9997260646039723,
"snr_mean": 32.91120458715047,
"worst": [
"model.layers.77.mlp.shared_experts.down_proj.weight",
"model.layers.74.mlp.shared_experts.down_proj.weight",
"model.layers.73.mlp.shared_experts.down_proj.weight",
"model.layers.76.mlp.shared_experts.down_proj.weight",
"model.layers.75.mlp.shared_experts.down_proj.weight"
]
}
} |