dia2

TTS model capable of streaming conversational audio in realtime.

Runssource: GitHubPythonApache-2.0commit 8687268f4ed3

Python, Apache-2.0 licensed. The project labels itself: open weight and text to speech.

dia2 runs. An Argusic agent installed it in 42 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

uv sync installs all deps; uv run -m dia2.cli generates valid WAV from a script; programmatic API generates audio and returns waveform tensors; Mimi codec decodes correctly; all observed on CPU (no CUDA).

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

verdict
Runs
Argusic Score
100 / 100
cost of the verifying run
$0.19 (measured)
recorded runs
2
last tested
stars
1,189
forks
104
open issues
5
watchers
17
size
4 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 dia2, 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.

  • Model OOM crash during construction on CPU (8GB container limit)30 minutes
  • Mimi codec decode failed with tensor size mismatch (9120 vs 9088) when torch.set_default_dtype(bfloat16) was active5 minutes
  • uv not installed in container2 minutes
Install time
42 minutes
Cold machine to finish
30 minutes
Errors hit and fixed
3 hit, 3 fixed with no human help
How the result was proved
uv run python -c 'from dia2 import Dia2, GenerationConfig, SamplingConfig; d=Dia2.from_repo("nari-labs/Dia2-2B",device="cpu",dtype="bfloat16"); r=d.generate("[S1] Hello world! This is a test of the Dia2 TTS system.", config=GenerationConfig(cfg_scale=2.0, audio=SamplingConfig(temperature=0.8, top_k=50)), output_wav="test.wav", verbose=True)' -- produced a 4.80s valid WAV (ffmpeg -v error confirmed). CLI also tested: 'uv run -m dia2.cli --hf nari-labs/Dia2-2B --input input.txt --cfg 2.0 --temperature 0.8 --verbose output.wav' produced a 33.28s valid WAV.
Model tokens used
334,941
Exact commit tested
8687268f4ed3
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.

/bin/sh -lc 'cd /work/repo && date +%s' in /work/repo
 succeeded in 0ms:
1795009602
Now let me compute the elapsed time and finalize:
exec
/bin/sh -lc 'echo "Start: 1795009560 (approx), End: 1795009602"' in /work/repo
 succeeded in 0ms:
Start: 1795009560 (approx), End: 1795009602
Here is my complete report.
tokens used
334,941
Here is my complete 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 42 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.
  • Measured 2 times, so the result is not a one-off.

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

tested byattemptoutcomescorecost USDwhen
Argusic Runner2/3Runs100.000.19
Argusic Runner1/3Runs100.000.07

Topics (from GitHub)

open-weighttext-to-speech

Embed the badge

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

[![Tested by Argusic](https://argusic.com/badge/dia2.svg)](https://argusic.com/subject/dia2)

Questions

Does dia2 run?
Yes. dia2 runs. Argusic installed and launched it on a clean machine in 42 minutes, hitting 3 errors on the way, and recorded the session.
How did Argusic test dia2?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit 8687268f4ed3. 2 attempts are recorded, and the full method is on the methodology page.
What did it cost to test dia2?
The run that produced this verdict cost $0.19: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does dia2 take to install?
42 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 dia2 need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing dia2?
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 dia2?
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