Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Seeed-Projects/Seeed-Jetson-DevelopTool --skill voice-llm-reachy-mini-physicalgit clone --depth 1 https://github.com/Seeed-Projects/Seeed-Jetson-DevelopToolWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/seeed-projects/seeed-jetson-developtool/voice-llm-reachy-mini-physical)<a href="https://agentmods.dev/skills/seeed-projects/seeed-jetson-developtool/voice-llm-reachy-mini-physical"><img src="https://agentmods.dev/badge/skills/seeed-projects/seeed-jetson-developtool/voice-llm-reachy-mini-physical/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/seeed-projects/seeed-jetson-developtool/voice-llm-reachy-mini-physical"><img src="https://agentmods.dev/badge/skills/seeed-projects/seeed-jetson-developtool/voice-llm-reachy-mini-physical.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00069 | $0.00877 |
| Opus 5 | $0.00034 | $0.00439 |
| Sonnet 5 | $0.00014 | $0.00175 |
| Haiku 4.5 | $0.00007 | $0.00088 |
Grade C, and why
voice-llm-reachy-mini-physical scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsSL https://ollama.com/install.sh | sh Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsSL https://ollama.com/install.sh | sh This is a copy
92% identical to voice-llm-reachy-mini-multimodal — 8 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Local Voice LLM on reComputer Mini for Reachy Mini (Physical AI)
Build a low-latency, privacy-first voice assistant on reComputer Mini J501 paired with Reachy Mini robot. Uses local speech recognition, Ollama LLM, and speech synthesis for embodied physical AI interaction.
Execution model
Run one phase at a time. After each phase:
- Relay all output to the user.
- If output contains
[STOP]→ stop immediately, consult the failure decision tree. - If output ends with
[OK]→ tell the user "Phase N complete" and proceed to the next phase.
Prerequisites
| Requirement | Detail |
|---|---|
| Jetson device | reComputer Mini J501 Kit (carrier board + Jetson module + cooling) |
| Robot | Reachy Mini Lite, connected via USB Type-A to J501 |
| JetPack | 6.2 |
| Network | Internet access for model downloads and repo cloning |
Phase 1 — Install Ollama and pull LLM (~15 min)
curl -fsSL https://ollama.com/install.sh | sh
ollama pull llama3.2-vision:11b
The model download takes approximately 10 minutes.
[OK] when ollama pull completes. [STOP] if install fails or download errors out.
Phase 2 — Install conversation application (~5 min)
cd Downloads
git clone https://github.com/Seeed-Projects/reachy-mini-loacl-conversation.git
cd reachy-mini-loacl-conversation
pip install -r requirements.txt -i https://pypi.jetson-ai-lab.io/
pip install "reachy-mini"
[OK] when all pip installs complete without error. [STOP] if dependency installation fails.
Phase 3 — Launch the application (~2 min)
In one terminal, start the Reachy Mini daemon:
reachy-mini-daemon
In a second terminal, set environment variables and launch:
export OLLAMA_HOST="http://localhost:11434"
export OLLAMA_MODEL="qwen2.5:7b"
export COQUI_MODEL_NAME="tts_models/zh-CN/baker/tacotron2-DDC-GST"
export DEFAULT_VOLUME="1.5"
python main.py
Use R key to start recording and S key to stop. The LLM will generate a spoken response.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 101 lines · 69 tokens per session scan C bc41e09b2d83
voice-llm-reachy-mini-physical is a skill published in the GitHub repository Seeed-Projects/Seeed-Jetson-DevelopTool (55 stars, last pushed yesterday), licensed MIT. It adds 69 tokens to every session and 877 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). It is 92% identical to voice-llm-reachy-mini-multimodal, differing in 8 lines, and is treated as a copy.
Other skills, from other repositories
spark-environment-setup
Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.
spark-memory-thermal-ops
Manage unified memory and thermals during long-running ML jobs on NVIDIA DGX Spark. Use when planning memory headroom for a training run on GB10, when a job OOMs on unified memory, or when monitoring temperature and power during multi-hour training.
spark-training-gotchas
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.
llama-cpp
Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.
minicpm5-deploy-vllm-ascend
Deploy MiniCPM5-2B with vLLM on Huawei Ascend NPU using vLLM-Ascend. Use when the user mentions vLLM-Ascend, Ascend NPU, Huawei Ascend, CANN, torchnpu, davinci devices, or wants an OpenAI-compatible MiniCPM5 server on Ascend hardware.
minicpm5-deploy-litert
Run MiniCPM5-2B or MiniCPM5-1B on-device with Google's LiteRT-LM runtime — the litert-lm CLI or its OpenAI-compatible server on a desktop, the Kotlin API or the AI Edge Gallery app on Android, the same .litertlm bundle on CPU or GPU. Use when the user says "LiteRT", "LiteRT-LM", "litertlm", ".litertlm", "Android"…