Text Generation
Safetensors
MLX
mlx-node
qwen3_5_moe
quantized
mxfp4
mxfp8
mixed-precision
apple-silicon
qwen3.5
Mixture of Experts
coding-agent
unsloth
conversational
3-bit
Instructions to use Brooooooklyn/Ornith-1.0-35B-mxfp4-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use Brooooooklyn/Ornith-1.0-35B-mxfp4-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("Brooooooklyn/Ornith-1.0-35B-mxfp4-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
- Unsloth Desktop
- Pi
How to use Brooooooklyn/Ornith-1.0-35B-mxfp4-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 "Brooooooklyn/Ornith-1.0-35B-mxfp4-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": "Brooooooklyn/Ornith-1.0-35B-mxfp4-mlx" } ] } } }Run Pi
# Start Pi in your project directory: pi
- MLX LM
How to use Brooooooklyn/Ornith-1.0-35B-mxfp4-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 "Brooooooklyn/Ornith-1.0-35B-mxfp4-mlx"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "Brooooooklyn/Ornith-1.0-35B-mxfp4-mlx" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Brooooooklyn/Ornith-1.0-35B-mxfp4-mlx", "messages": [ {"role": "user", "content": "Hello"} ] }' - Hermes Agent
How to use Brooooooklyn/Ornith-1.0-35B-mxfp4-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 "Brooooooklyn/Ornith-1.0-35B-mxfp4-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 Brooooooklyn/Ornith-1.0-35B-mxfp4-mlx
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use Brooooooklyn/Ornith-1.0-35B-mxfp4-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 "Brooooooklyn/Ornith-1.0-35B-mxfp4-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 "Brooooooklyn/Ornith-1.0-35B-mxfp4-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"
| { | |
| "schemaVersion": 1, | |
| "modelPath": "/Users/brooklyn/workspace/github/mlx-node/.cache/models/ornith-1.0-35b-unsloth-mxfp4-mlx", | |
| "startedAt": "2026-07-20T16:43:42.832Z", | |
| "completedAt": "2026-07-20T16:54:10.310Z", | |
| "environment": { | |
| "platform": "darwin", | |
| "arch": "arm64", | |
| "osRelease": "25.5.0", | |
| "osVersion": "Darwin Kernel Version 25.5.0: Tue Jun 9 22:28:34 PDT 2026; root:xnu-12377.121.10~1/RELEASE_ARM64_T6050", | |
| "cpu": "Apple M5 Max", | |
| "logicalCpus": 18, | |
| "totalMemoryBytes": 137438953472, | |
| "nodeVersion": "v24.13.1", | |
| "packages": { | |
| "lm": { | |
| "name": "@mlx-node/lm", | |
| "version": "0.0.8" | |
| }, | |
| "core": { | |
| "name": "@mlx-node/core", | |
| "version": "0.0.8" | |
| }, | |
| "native": { | |
| "name": "@mlx-node/core-darwin-arm64", | |
| "version": "0.0.8" | |
| } | |
| } | |
| }, | |
| "configuration": { | |
| "runs": 3, | |
| "maxNewTokens": 512, | |
| "cooldownSeconds": 60, | |
| "warmupRuns": 0, | |
| "temperature": 0, | |
| "reasoningEffort": "none", | |
| "reportPerformance": true, | |
| "minimumCompletionRatio": 0.95, | |
| "systemPrompt": "You are a technical writer in a deterministic inference benchmark. Produce a continuous, detailed response and do not conclude early.", | |
| "prompt": "Write a comprehensive technical handbook chapter about building a production compiler. Cover lexical analysis, parsing, semantic analysis, type checking, intermediate representations, optimization, code generation, linking, testing, debugging, and deployment. Use detailed explanations and concrete examples. Do not summarize or conclude early; keep expanding the chapter until the generation limit is reached.", | |
| "isolation": "fresh child process and model load per sample" | |
| }, | |
| "warmups": [], | |
| "runs": [ | |
| { | |
| "pid": 4948, | |
| "loadMs": 295858.866667, | |
| "generationWallMs": 8295.485790999955, | |
| "totalWallMs": 304154.46933299996, | |
| "ttftMs": 473.89941699999997, | |
| "prefillTokensPerSecond": 223.6761561578372, | |
| "decodeTokensPerSecond": 67.44062557118689, | |
| "promptTokens": 106, | |
| "generatedTokens": 512, | |
| "reasoningTokens": 0, | |
| "cachedTokens": 0, | |
| "finishReason": "length", | |
| "minimumGeneratedTokens": 487, | |
| "phase": "measured", | |
| "run": 1 | |
| }, | |
| { | |
| "pid": 14146, | |
| "loadMs": 77110.606125, | |
| "generationWallMs": 6654.497124999994, | |
| "totalWallMs": 83765.210292, | |
| "ttftMs": 361.292125, | |
| "prefillTokensPerSecond": 293.39139346034733, | |
| "decodeTokensPerSecond": 82.95027113210845, | |
| "promptTokens": 106, | |
| "generatedTokens": 512, | |
| "reasoningTokens": 0, | |
| "cachedTokens": 0, | |
| "finishReason": "length", | |
| "minimumGeneratedTokens": 487, | |
| "phase": "measured", | |
| "run": 2 | |
| }, | |
| { | |
| "pid": 17338, | |
| "loadMs": 111887.36429200001, | |
| "generationWallMs": 6649.128375, | |
| "totalWallMs": 118536.607042, | |
| "ttftMs": 364.8845, | |
| "prefillTokensPerSecond": 290.5028851595505, | |
| "decodeTokensPerSecond": 83.18705959265836, | |
| "promptTokens": 106, | |
| "generatedTokens": 512, | |
| "reasoningTokens": 0, | |
| "cachedTokens": 0, | |
| "finishReason": "length", | |
| "minimumGeneratedTokens": 487, | |
| "phase": "measured", | |
| "run": 3 | |
| } | |
| ], | |
| "medians": { | |
| "loadMs": 111887.36429200001, | |
| "generationWallMs": 6654.497124999994, | |
| "totalWallMs": 118536.607042, | |
| "ttftMs": 364.8845, | |
| "prefillTokensPerSecond": 290.5028851595505, | |
| "decodeTokensPerSecond": 82.95027113210845, | |
| "promptTokens": 106, | |
| "generatedTokens": 512 | |
| } | |
| } | |