Get your rig on the board.

One command. About two minutes.

  1. Load a model

    Any model, any quant, loaded in any local server.

    Node 18 or newer

  2. Run the bench

    The CLI finds the server, reads which model and quant it is running, signs you in with GitHub the first time, runs a fixed test and posts your score.

    npx toks-gg bench
  3. Get your board

    Your rig lands on the board for that model and quant, with its rank and a share card.

    See the board

Bench whatever you actually run.

Each model and quant gets its own board, so you're always compared like with like.

Ollama

  1. Load the model you want to test (any model, any quant):

    ollama run qwen3:32b-q4_K_M "hi"
  2. Bench it:

    npx toks-gg bench

Bench options

Add --name "Your rig name" and --cost 1900 to show up in the $ per tok/s ranking.

Options for npx toks-gg bench
OptionWhat it does
--model <name>which loaded model to bench, if the server has several
--url <url>your server, if it isn't on a common port
--backend <name>pick one when several are running (ollama, llama.cpp, vllm, mlx, ...)
--tp <n>tensor-parallel size, for vLLM across several GPUs
--name <name>a name for your rig, e.g. "The Space Heater"
--cost <usd>what the rig cost you, for the $ per tok/s ranking
--dry-runrun it, don't submit

What gets sent

Nothing else. No prompts or files from your machine.

Everything a bench run sends to toks.gg
ItemExactly what is sent
GitHub usernameFrom the sign-in, shown on your rig page.
GPUsTheir names as nvidia-smi, rocm-smi or macOS report them, plus the PCIe link on NVIDIA.
CPUThe model name.
RAMTotal size in GB.
Operating systemIts name and version.
ServerThe backend's name, and the tensor-parallel size if you pass --tp.
ModelThe name the server gives it, plus its file name, quant and file size when the server reports them.
SpeedsGeneration and prompt processing, in tok/s.
Output hashA SHA-256 hash of the model's output, not the output itself.
Run detailsThe one-time challenge ID and the CLI version.
Rig name and costOnly when you pass --name or --cost.