NGT
Nearest Neighbor Search with Neighborhood Graph and Tree for High-dimensional Data
Runssource: GitHubC++Apache-2.0commit 573dab62b783
C++, Apache-2.0 licensed. The project labels itself: approximate nearest neighbor search, k nearest neighbors, knn search, nearest neighbor search, nearest neighbors, vector database and vector search.
NGT runs. An Argusic agent installed it in 4 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
NGT v2.8.1 C++ library, command-line tool, and four sample programs build and run successfully; the CLI creates indexes, appends data, searches with L2/cosine/hamming distance, and reports index statistics; the Python ctypes binding (ngt.base) inserts 5000 128-d vectors and returns correct nearest-neighbor results.
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, refreshed . 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.
- Python ngtpy C++ extension build failed: python3-dev (Python.h) not available on system and cannot be installed without root2 minutes
- Python ngtpy C++ extension build failed: missing generated headers defines.h and version_defs.h in include path1 minute
- make install: Permission denied writing to /usr/local/lib0.5 minutes
- Install time
- 4 minutes
- Cold machine to finish
- 11 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- How the result was proved
- ./build/bin/ngt/ngt create -d 128 -o c /tmp/final-test data/sift-dataset-5k.tsv (exit 0); ./build/bin/ngt/ngt search -n 3 /tmp/final-test data/sift-query-3.tsv (exit 0, returned 3 query results with correct distances); ngt info /tmp/final-test (exit 0, 5000 objects, 99890 edges); ./build/samples/{l2-uint8,cosine-float,hamming-uint8} all ran and printed search results; Python ctypes module: ngt.base.Index.insert_blob + Index.search returned matching results
- Model tokens used
- 106,510
- Exact commit tested
- 573dab62b783
- 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 'find /tmp/final-test -type f 2>/dev/null' in /work/repo succeeded in 0ms: /tmp/final-test/obj /tmp/final-test/robj /tmp/final-test/grp /tmp/final-test/prf /tmp/final-test/tre Now I have all the information needed for the final report. tokens used 106,510 Now I have all the information needed for the final report.
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 4 minutes.
- Recovered from all 3 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.
- A quick evaluation: it was running 4 minutes after the clone.
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)
approximate-nearest-neighbor-searchk-nearest-neighborsknn-searchnearest-neighbor-searchnearest-neighborsvector-databasevector-search
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/ngt)Questions
- Does NGT run?
- Yes. NGT runs. Argusic installed and launched it on a clean machine in 4 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test NGT?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 573dab62b783. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test NGT?
- The run that produced this verdict cost $0.08: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does NGT take to install?
- 4 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 NGT need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing NGT?
- 3 things broke in the recorded run, and 3 were fixed without human help. Each one, and the time it cost, is listed on this page.
- Where is the evidence for NGT?
- 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.