vllm-mlx runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 12 minutes, hitting 1 error on the way, and recorded the session.

How did Argusic test vllm-mlx?

On a fresh, disposable machine, with every command recorded and the repository pinned at commit 37a16c76bb21. One attempt is recorded, and the full method is on the methodology page.

What did it cost to test vllm-mlx?

The run that produced this verdict cost $0.10: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.

How long does vllm-mlx take to install?

12 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 vllm-mlx need a GPU?

Not to start. Argusic reached a running state on a machine with no GPU.

What goes wrong when installing vllm-mlx?

1 thing broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.

Where is the evidence for vllm-mlx?

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.

vllm-mlx

High-performance OpenAI and Anthropic compatible LLM inference server for Apple Silicon. Native MLX, continuous batching, multimodal models, MCP tool calling, and Claude Code support.

Runs with mockssource: GitHub homepagePythonApache-2.0commit 37a16c76bb21

Python, Apache-2.0 licensed. The project labels itself: anthropic, anthropic api, apple silicon, claude code, continuous batching, inference server, llm and local llm.

vllm-mlx runs, with stand-ins for the services it depends on. An Argusic agent installed it in 12 minutes and hit 1 error and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

vllm-mlx 0.5.0 installs into a venv on x86 Linux, its CLI help command prints usage for 8 subcommands, and 1559 of its Linux-compatible unit tests pass
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 92 of 100 (the mean of this project's run scores).

At a glance

verdict: Runs with mocks
Argusic Score: 92 / 100
cost of the verifying run: $0.10 (measured)
recorded runs: 1
last tested: 2026-10-06
stars: 1,611
forks: 229
open issues: 117
watchers: 10
size: 14 MB
created: 2025-12-06
last push: 2026-10-03

Subject data from GitHub, linked at the top of this page, refreshed 2026-10-06. 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.

Time lost to each failure while testing vllm-mlx, drawn from the recorded run — Every bar is a failure the agent recorded, with the time it cost. Drawn from this run, not from anywhere else.

What went wrong, in the agent's own words

The agent's own words, worst first, never rewritten into advice.

4 Apple-Silicon-only test scripts (test_paged_cache_benefits.py, test_paged_cache_real_inference.py, test_paged_cache_real_model.py, test_registry_idle_unload_real_model.py) called sys.exit(0) at module level on non-Apple-Silicon platforms, causing pytest collection to fail with SystemExit for all 11 attempted collection-errors3 minutes

Install time: 12 minutes
Cold machine to finish: 23 minutes
Errors hit and fixed: 1 hit, 1 fixed with no human help
How the result was proved: pytest ran 1553 passed (Linux CI test matrix) plus 6 passed (MLX-stubbed scheduler tests), 15 deselected, 0 failed
Model tokens used: 117,723
Exact commit tested: 37a16c76bb21
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
Behaviour against real services: this run was verified against stand-in services only

The end of the recording

The last lines of the terminal session that decided this verdict, exactly as they were recorded.

    if "pytest" in sys.modules:
        import pytest
        pytest.skip("This test requires Apple Silicon", allow_module_level=True)
    else:
        print("This test requires Apple Silicon")
        sys.exit(0)
def print_header(title: str) -> None:
    """Print a formatted header."""
    print("\n" + "=" * 70)
    print(f"  {title}")
tokens used
117,723

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

Recovered from all 1 error without a human stepping in, which says the failures are documented well enough to solve.
Licensed Apache-2.0, as reported by its host.

What did not, or is not known

Only reached a running state against stand-in services, so nothing here proves it works against the real ones.
Took 12 minutes to install, slower than the median of the 3 comparable projects Argusic has measured.
Measured once so far. A single attempt shows it can run, not that it runs reliably.

Of the 4 anthropic projects Argusic has installed and timed, vllm-mlx was the 3rd fastest to reach a running state, and 2 of 4 reached one at all.

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

tested by	attempt	outcome	score	cost USD	when
Argusic Runner	1/3	Runs with mocks	92.00	0.10	2026-10-06 16:40 UTC

Embed the badge

Markdown for the project README. It links back here; terms on the terms page.

[![Tested by Argusic](https://argusic.com/badge/vllm-mlx.svg)](https://argusic.com/subject/vllm-mlx)

Questions

Does vllm-mlx run?: vllm-mlx runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 12 minutes, hitting 1 error on the way, and recorded the session.
How did Argusic test vllm-mlx?: On a fresh, disposable machine, with every command recorded and the repository pinned at commit 37a16c76bb21. One attempt is recorded, and the full method is on the methodology page.
What did it cost to test vllm-mlx?: The run that produced this verdict cost $0.10: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does vllm-mlx take to install?: 12 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 vllm-mlx need a GPU?: Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing vllm-mlx?: 1 thing broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.
How does vllm-mlx compare with the alternatives?: Of the 4 anthropic projects Argusic has installed and timed, vllm-mlx was the 3rd fastest to reach a running state, and 2 of 4 reached one at all.
Where is the evidence for vllm-mlx?: 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.