# Send MLX Local Inference Stack to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "mlx-local-inference",
    "name": "MLX Local Inference Stack",
    "source": "tencent",
    "type": "skill",
    "category": "通讯协作",
    "sourceUrl": "https://clawhub.ai/bendusy/mlx-local-inference",
    "canonicalUrl": "https://clawhub.ai/bendusy/mlx-local-inference",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/mlx-local-inference",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=mlx-local-inference",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "README.md",
      "README_CN.md",
      "SKILL.md",
      "references/asr-qwen3.md",
      "references/asr-whisper.md",
      "references/embedding-qwen3.md"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-23T16:43:11.935Z",
      "expiresAt": "2026-04-30T16:43:11.935Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
        "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.zip\"",
        "redirectLocation": null,
        "bodySnippet": null
      },
      "scope": "source",
      "summary": "Source download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this source.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/mlx-local-inference"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/mlx-local-inference",
    "downloadUrl": "https://openagent3.xyz/downloads/mlx-local-inference",
    "agentUrl": "https://openagent3.xyz/skills/mlx-local-inference/agent",
    "manifestUrl": "https://openagent3.xyz/skills/mlx-local-inference/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/mlx-local-inference/agent.md"
  }
}
```
## Documentation

### MLX Local Inference Stack

Full local AI inference on Apple Silicon Macs. All services expose OpenAI-compatible APIs.

### Services Overview

ServicePortAccessModelsLLM + Whisper + Embedding8787LAN (0.0.0.0)qwen3-14b, gemma-3-12b, whisper-large-v3-turbo, qwen3-embedding-0.6b/4bASR (Qwen3-ASR)8788localhost onlyQwen3-ASR-1.7B-8bitTranscribe Daemon—file-basedUses ASR + LLM

LaunchAgents: com.mlx-server (8787), com.mlx-audio-server (8788), com.mlx-transcribe-daemon

### Models

Model IDParamsBest Forqwen3-14b14B 4bitChinese, deep reasoning (built-in think mode)gemma-3-12b12B 4bitEnglish, code generation

### API

curl -X POST http://localhost:8787/v1/chat/completions \\
  -H "Content-Type: application/json" \\
  -d '{
    "model": "qwen3-14b",
    "messages": [{"role": "user", "content": "Hello"}],
    "temperature": 0.7,
    "max_tokens": 2048
  }'

Add "stream": true for streaming.

### Python

from openai import OpenAI
client = OpenAI(base_url="http://localhost:8787/v1", api_key="unused")
response = client.chat.completions.create(
    model="qwen3-14b",
    messages=[{"role": "user", "content": "Hello"}],
    temperature=0.7, max_tokens=2048
)
print(response.choices[0].message.content)

### Qwen3 Think Mode

Qwen3 may include <think>...</think> chain-of-thought tags. Strip them:

import re
text = re.sub(r'<think>.*?</think>\\s*', '', text, flags=re.DOTALL)

### Model Selection Guide

ScenarioRecommendedChinese textqwen3-14bCantoneseqwen3-14bEnglish writinggemma-3-12bCode generationEitherDeep reasoningqwen3-14b (think mode)Quick Q&Agemma-3-12b

### Qwen3-ASR (best for Chinese/Cantonese)

curl -X POST http://127.0.0.1:8788/v1/audio/transcriptions \\
  -F "file=@audio.wav" \\
  -F "model=mlx-community/Qwen3-ASR-1.7B-8bit" \\
  -F "language=zh"

### Whisper (multilingual, 99 languages)

curl -X POST http://localhost:8787/v1/audio/transcriptions \\
  -F "file=@audio.wav" \\
  -F "model=whisper-large-v3-turbo"

### ASR Model Comparison

Qwen3-ASR (port 8788)Whisper (port 8787)Chinese/CantoneseStrongAverageMultilingualNoYes (99 langs)LAN accessNo (localhost)YesLoadingOn-demandAlways loaded

### Supported audio formats

wav, mp3, m4a, flac, ogg, webm

### Long audio

Split into 10-min chunks first:

ffmpeg -y -ss 0 -t 600 -i long.wav -ar 16000 -ac 1 chunk_000.wav

### Models

Model IDSizeUse Caseqwen3-embedding-0.6b0.6B 4bitFast retrieval, low latencyqwen3-embedding-4b4B 4bitHigh-accuracy semantic matching

### API

curl -X POST http://localhost:8787/v1/embeddings \\
  -H "Content-Type: application/json" \\
  -d '{"model": "qwen3-embedding-0.6b", "input": "text to embed"}'

### Batch

curl -X POST http://localhost:8787/v1/embeddings \\
  -H "Content-Type: application/json" \\
  -d '{"model": "qwen3-embedding-4b", "input": ["text 1", "text 2"]}'

### Default Model: PaddleOCR-VL-1.5-6bit

ItemValueModel IDpaddleocr-vl-6bitSpeed~185 t/sMemory~3.3 GBPromptOCR:

### CLI

cd ~/.mlx-server/venv
python -m mlx_vlm.generate \\
  --model mlx-community/PaddleOCR-VL-1.5-6bit \\
  --image image.jpg \\
  --prompt "OCR:" \\
  --max-tokens 512 --temp 0.0

### Python

from mlx_vlm import generate, load
from mlx_vlm.prompt_utils import apply_chat_template
from mlx_vlm.utils import load_config

model, processor = load("mlx-community/PaddleOCR-VL-1.5-6bit")
config = load_config("mlx-community/PaddleOCR-VL-1.5-6bit")
prompt = apply_chat_template(processor, config, "OCR:", num_images=1)
out = generate(model, processor, prompt, "image.jpg",
               max_tokens=512, temperature=0.0, verbose=False)
print(out.text if hasattr(out, "text") else out)

### Notes

Prompt must be exactly OCR: for PaddleOCR-VL
temperature=0.0 for deterministic output
RGBA images must be converted to RGB first
Venv: ~/.mlx-server/venv

### Model: Qwen3-TTS (cached, not auto-served)

ItemValueModelQwen3-TTS-12Hz-1.7B-CustomVoice-8bitMemory~2GBFeatureCustom voice cloning

### CLI

~/.mlx-server/venv/bin/mlx_audio.tts.generate \\
  --model mlx-community/Qwen3-TTS-12Hz-1.7B-CustomVoice-8bit \\
  --text "你好，这是一段测试语音"

### As API (via mlx_audio.server on port 8788)

curl -X POST http://127.0.0.1:8788/v1/audio/speech \\
  -H "Content-Type: application/json" \\
  -d '{
    "model": "mlx-community/Qwen3-TTS-12Hz-1.7B-CustomVoice-8bit",
    "input": "你好世界"
  }' --output speech.wav

### 6. Transcribe Daemon — Automatic Batch Transcription

Drop audio files into ~/transcribe/ for automatic processing:

Daemon detects file (polls every 15s)
Phase 1: Transcribe via Qwen3-ASR → filename_raw.md
Phase 2: Correct via Qwen3-14B LLM → filename_corrected.md
Move results to ~/transcribe/done/

### LLM Correction Rules

Fix homophone errors (的/得/地, 在/再)
Preserve Cantonese characters (嘅、唔、咁、喺、冇、佢)
Add punctuation and paragraphs
Remove filler words

### Supported formats

wav, mp3, m4a, flac, ogg, webm

### Service Management

# LLM + Whisper + Embedding server (port 8787)
launchctl kickstart -k gui/$(id -u)/com.mlx-server

# ASR server (port 8788)
launchctl kickstart -k gui/$(id -u)/com.mlx-audio-server

# Transcribe daemon
launchctl kickstart gui/$(id -u)/com.mlx-transcribe-daemon

# Logs
tail -f ~/.mlx-server/logs/server.log
tail -f ~/.mlx-server/logs/mlx-audio-server.err.log
tail -f ~/.mlx-server/logs/transcribe-daemon.err.log

### Requirements

Apple Silicon Mac (M1/M2/M3/M4)
Python 3.10+ with mlx, mlx-lm, mlx-audio, mlx-vlm
Recommended: 32GB+ RAM for running multiple models
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: bendusy
- Version: 2.2.0
## Source health
- Status: healthy
- Source download looks usable.
- Yavira can redirect you to the upstream package for this source.
- Health scope: source
- Reason: direct_download_ok
- Checked at: 2026-04-23T16:43:11.935Z
- Expires at: 2026-04-30T16:43:11.935Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/mlx-local-inference)
- [Send to Agent page](https://openagent3.xyz/skills/mlx-local-inference/agent)
- [JSON manifest](https://openagent3.xyz/skills/mlx-local-inference/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/mlx-local-inference/agent.md)
- [Download page](https://openagent3.xyz/downloads/mlx-local-inference)