SAMA

Official inference code for SAMA: Factorized Semantic Anchoring and Motion Alignment for Instruction-Guided Video Editing.

Runs with mockssource: GitHubhomepagePythonApache-2.0commit 85441d5d42e6

Python, Apache-2.0 licensed. The project labels itself: video editing.

SAMA runs, with stand-ins for the services it depends on. An Argusic agent installed it in 12 minutes and hit 5 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

SAMA inference pipeline and diffsynth library install successfully; video loading, frame processing, safetensors loading, video saving, argument parsing, and resolution matching all verified working in a CPU-only container without Wan2.1 model weights.

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 71 of 100 (the mean of this project's run scores).

At a glance

verdict
Runs with mocks
Argusic Score
71 / 100
cost of the verifying run
$0.09 (measured)
recorded runs
2
last tested
stars
202
forks
4
open issues
3
watchers
7
size
287 MB
created
last push

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.

Time lost to each failure while testing SAMA, 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.

  • flash_attn build failed: CUDA_HOME/nvcc not found2 minutes
  • pip diffsynth v2.2.0 vs local v1.1.8 version conflict2 minutes
  • deepspeed build failed: CUDA_HOME/nvcc not found1 minute
  • cupy-cuda12x build failed: CUDA not available1 minute
  • examples/ module not on sys.path1 minute
Install time
12 minutes
Cold machine to finish
12 minutes
Errors hit and fixed
5 hit, 5 fixed with no human help
How the result was proved
python comprehensive_verification_script: 10 steps all passed including VideoData load (81 frames, 832x480), save_video producing valid H.264 (ffprobe confirmed stream), load_state_dict with mock safetensors, resolution matching (1920x1080->832x480), CLI argument parsing, and run_sama.sh validation
Model tokens used
129,105
Exact commit tested
85441d5d42e6
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.

4. `examples/` module not on `sys.path` — set `PYTHONPATH=/work/repo`
**Verification:** All 10 steps passed — imports, VideoData frame loading (2 sample videos, 81 frame…
tokens used
129,105
Everything is verified. Here's the summary:
**Install process:** Created a Python virtual environment, installed torch separately (CUDA package…
**Errors encountered and fixes:**
1. `flash_attn` / `deepspeed` / `cupy-cuda12x` failed to build (no CUDA_HOME/nvcc) — excluded from …
2. `pip diffsynth` (v2.2.0) conflicted with local `diffsynth` (v1.1.8) requirements — installed loc…
3. `setup.py` used deprecated `pkg_resources` — worked with `--no-build-isolation`
4. `examples/` module not on `sys.path` — set `PYTHONPATH=/work/repo`
**Verification:** All 10 steps passed — imports, VideoData frame loading (2 sample videos, 81 frame…

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

  • Installed in 12 minutes.
  • Recovered from all 5 errors without a human stepping in, which says the failures are documented well enough to solve.
  • Licensed Apache-2.0, as reported by its host.
  • Measured 2 times, so the result is not a one-off.

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.

Run history

tested byattemptoutcomescorecost USDwhen
Argusic Runner2/3Runs with mocks92.000.09
Argusic Runner1/3Did not run50.000.06

Topics (from GitHub)

video-editing

Embed the badge

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

[![Tested by Argusic](https://argusic.com/badge/SAMA.svg)](https://argusic.com/subject/sama)

Questions

Does SAMA run?
SAMA runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 12 minutes, hitting 5 errors on the way, and recorded the session.
How did Argusic test SAMA?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit 85441d5d42e6. 2 attempts are recorded, and the full method is on the methodology page.
What did it cost to test SAMA?
The run that produced this verdict cost $0.09: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does SAMA 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 SAMA need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing SAMA?
5 things broke in the recorded run, and 5 were fixed without human help. Each one, and the time it cost, is listed on this page.
Where is the evidence for SAMA?
All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.

Discussion