{"_id":"@altronis/tokenmark-cli","name":"@altronis/tokenmark-cli","dist-tags":{"latest":"0.1.0"},"versions":{"0.1.0":{"name":"@altronis/tokenmark-cli","version":"0.1.0","description":"TokenMark CLI — recommend the best local-LLM model + config for your hardware from real benchmarks (Strix Halo, DGX Spark, RTX, and more).","type":"module","bin":{"tokenmark":"bin/tokenmark.mjs"},"engines":{"node":">=18"},"keywords":["llm","local-llm","benchmark","strix-halo","dgx-spark","gguf","tokens-per-second","llama.cpp","tokenmark"],"repository":{"type":"git","url":"git+https://github.com/sypherin/TokenMark.git"},"homepage":"https://tokenmark.app","author":{"name":"Altronis"},"license":"MIT","publishConfig":{"access":"public"},"_id":"@altronis/tokenmark-cli@0.1.0","gitHead":"4372291369769bb6977783fa7405255550e48e7f","bugs":{"url":"https://github.com/sypherin/TokenMark/issues"},"_nodeVersion":"22.22.2","_npmVersion":"10.9.7","dist":{"integrity":"sha512-TLTZzrXTlIkclk8A/emmfg5ojiL045d0SpySCUCTDtk3DF5ciqLJuY11L3cEDCe54tsMYUSWenwrBcHVDOoF+g==","shasum":"33bd9e623d1adb03d13c07d35ebdce7020102f8d","tarball":"https://registry.npmjs.org/@altronis/tokenmark-cli/-/tokenmark-cli-0.1.0.tgz","fileCount":4,"unpackedSize":12638,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEUCIQCkBeS4OsAQzqiUWSPakB6lVbY/kqjnnrJL5siEhV/niwIgbPaIvX7SW7NxLG6H+p3a+tYBgdVue2NBjbvZSWx99mI="}]},"_npmUser":{"name":"altronis","email":"sypherin@gmail.com"},"directories":{},"maintainers":[{"name":"altronis","email":"sypherin@gmail.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/tokenmark-cli_0.1.0_1788058401412_0.0875912451623766"},"_hasShrinkwrap":false}},"time":{"created":"2026-08-30T02:53:21.267Z","0.1.0":"2026-08-30T02:53:21.538Z","modified":"2026-08-30T02:53:21.727Z"},"maintainers":[{"name":"altronis","email":"sypherin@gmail.com"}],"description":"TokenMark CLI — recommend the best local-LLM model + config for your hardware from real benchmarks (Strix Halo, DGX Spark, RTX, and more).","homepage":"https://tokenmark.app","keywords":["llm","local-llm","benchmark","strix-halo","dgx-spark","gguf","tokens-per-second","llama.cpp","tokenmark"],"repository":{"type":"git","url":"git+https://github.com/sypherin/TokenMark.git"},"author":{"name":"Altronis"},"bugs":{"url":"https://github.com/sypherin/TokenMark/issues"},"license":"MIT","readme":"# @altronis/tokenmark-cli\n\nRecommend the best local-LLM **model + config** for your hardware, from real community benchmarks — [TokenMark](https://tokenmark.app) in your terminal.\n\nIt pulls the live benchmark snapshot from tokenmark.app and ranks **measured** configs (Strix Halo, DGX Spark, RTX, Mac). It never invents numbers — every pick is a real config with a source link.\n\n## Use\n\n```sh\nnpx @altronis/tokenmark-cli recommend --hardware \"Strix Halo\" --tasks coding,reasoning --prefer balanced\n```\n\nOr install it:\n\n```sh\nnpm i -g @altronis/tokenmark-cli\ntokenmark recommend --hardware \"DGX Spark\" --tasks vision --prefer quality\n```\n\n## Commands\n\n```\ntokenmark recommend --hardware <hw> [--tasks a,b] [--prefer speed|quality|balanced] [--limit N]\ntokenmark configs   [--model <name>] [--hardware <hw>]\ntokenmark search    <term>\ntokenmark models\n```\n\n- `--tasks`: any of `coding`, `reasoning`, `long-context`, `vision`, `agentic`, `general`\n- `--prefer`: `speed` (fastest), `quality` (most capable), `balanced` (default)\n\nPoint it at a different source with `TOKENMARK_URL` (default `https://tokenmark.app`).\n\n## What it shows\n\nEach pick is a real measured config: decode tok/s, quant, backend, the hardware it ran on, why it was picked, who benchmarked it, and a link to the source. Aggregate/multi-GPU numbers are labeled so they aren't read as single-stream speed.\n\nMIT licensed. Data aggregated from public community benchmarks with attribution.\n","readmeFilename":"README.md","_rev":"1-f0764abb3bcb29c614c01fb5ff236148"}