FreeToken
FreeToken brings datacenter-scale model serving to your desktop. Run massive models locally, fast and efficiently.
Runssource: GitHubPythonApache-2.0commit 0d652e73a452
Python, Apache-2.0 licensed. The project labels itself: deepseek v4, edge, glm, inference, minimax, moe, qwen and transformer.
FreeToken runs. An Argusic agent installed it in 32 minutes and hit 4 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
freetoken 0.1.3 (PyPI wheel + repo source overlay) installs and runs on CPU-only Linux x86_64 (no GPU); the ft CLI is functional, all 901 non-GPU tests pass, and all GPU-dependent tests correctly skip.
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 100 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page. Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.
What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- CUDA_HOME and CUDA toolkit not found; setup.py hard-requires it for C++ extensions (pinned_tensor, cpu_moe)5 minutes
- PyPI wheel (0.1.3) is missing ROCm arch functions (is_rocm, get_rocm_gfx_arch, _rocm_link_flags) that the repo source has and tests expect5 minutes
- ninja binary not on PATH; tvm_ffi subprocess calls fail with FileNotFoundError when JIT-compiling radix C++ kernel2 minutes
- python3-dev headers missing; cannot compile row_store C++ extension from repo source2 minutes
- Install time
- 32 minutes
- Cold machine to finish
- 28 minutes
- Errors hit and fixed
- 4 hit, 4 fixed with no human help
- How the result was proved
- pytest on tests/server/ (559 passed), tests/tokenizer/ (27 passed), tests/engine/ (112 passed), tests/daemon/ (31 passed), tests/checkpoint/ (17 passed), tests/utils/ (5 passed), tests/kernels/ -m 'not slow' (30 passed), tests/scheduler/ -m 'not slow' (99 passed), tests/layers/ tests/mm/ tests/attention/ (21 passed). ft --version prints 'freetoken version 0.1.3'. All key module imports verified (ServerArgs, Req, SamplingParams, CpuMoeExecutor, create_pinned_tensor_like).
- Model tokens used
- 851,256
- Exact commit tested
- 0d652e73a452
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
exec /bin/sh -lc 'cd /work/repo && export PATH="/home/runner/.codex/tmp/arg0/codex-arg0HRr2xD:$PATH" && … succeeded in 5787ms: 5 passed in 4.26s exec /bin/sh -lc 'echo "Elapsed: approximately 32 minutes based on wall clock observation"' in /work/repo succeeded in 0ms: Elapsed: approximately 32 minutes based on wall clock observation Now I have all the information. Here's the final result: tokens used 851,256 Now I have all the information. Here's the final result:
Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Installed in 32 minutes.
- Recovered from all 4 errors without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed Apache-2.0, as reported by its host.
What did not, or is not known
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
Topics (from GitHub)
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/freetoken)Questions
- Does FreeToken run?
- Yes. FreeToken runs. Argusic installed and launched it on a clean machine in 32 minutes, hitting 4 errors on the way, and recorded the session.
- How did Argusic test FreeToken?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 0d652e73a452. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test FreeToken?
- The run that produced this verdict cost $0.20: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does FreeToken take to install?
- 32 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does FreeToken need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing FreeToken?
- 4 things broke in the recorded run, and 4 were fixed without human help. Each one, and the time it cost, is listed on this page.
- Where is the evidence for FreeToken?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.