technode-cli 0.3.0

TechNode CLI — run inference on compressed open models, rent GPUs, or serve as a provider (RunPod-compatible).

technode

Run inference on TechNode — a GPU grid serving compressed open models (Qwen, Granite, gpt-oss, Devstral, Gemma, EXAONE…) at consumer-GPU prices.

pip install technode-cli       # the command is `technode`
technode login                 # paste your tn_test_… key (or set TECHNODE_API_KEY)
technode models                # list the compressed catalog
technode infer "Explain quantization in one line." --model qwen2.5-7b

Zero dependencies — pure Python stdlib, runs anywhere Python ≥3.8 does.

Commands

Command	What it does
technode login [key]	Save your API key to ~/.technode/config.json (chmod 600).
technode logout	Remove the saved key.
technode models [--json]	List available models (id, quantization, role).
technode infer PROMPT [-m MODEL] [-n MAX_TOKENS] [-t TEMP] [--json] [-q]	Text generation. - or piped stdin reads the prompt from stdin.
technode whoami	Show the active key (masked) + endpoint.
technode gpu lease/list/status/release	Rent a whole GPU (Jupyter lab session).

Become a provider (share your GPU)

Got an NVIDIA Linux box? Join the grid and serve models — outbound-only, works behind any NAT (no Tailscale, no inbound ports):

technode provider register --gpu "RTX 4090" --vram 24
technode provider serve --llama-server /path/to/llama-server   # pull-mode worker (llama.cpp)
technode provider status

Data-center / IDC GPUs (B200·B300·H100, multi-GPU)

For datacenter GPU nodes, use the vLLM backend with tensor-parallel across GPUs:

pip install -U technode-cli vllm        # vLLM needs CUDA GPUs + drivers
technode provider register --gpu "8x B200" --vram 1440
technode provider serve --backend vllm --tp 8     # 8-GPU tensor-parallel
technode provider install --backend vllm --tp 8   # → systemd unit (boot persistence)

One-shot onboarding (detects GPUs, installs, registers, serves):

curl -fsSL https://technode.network/idc.sh | bash -s -- --name "IDC-node-01"

serve polls the broker for jobs it can run, executes them on your GPU, and returns the results. Needs a llama.cpp llama-server binary (CUDA build for NVIDIA) and operator approval before it receives live jobs.

Configuration

Setting	Env var	Default
API key	TECHNODE_API_KEY	— (from technode login)
Endpoint	TECHNODE_BASE_URL	https://technode.network

Get a key (free beta): https://technode.network/developers

Examples

# pick a coder model
technode infer "Write a Python one-liner to flatten a list of lists." -m qwen2.5-coder-7b

# read the prompt from a file / pipe
cat prompt.txt | technode infer -

# machine-readable
technode infer "hi" --json