paritok-4b-v1
Non-destructive compression gateway for AI coding agents. Cuts token bills 25% on turn 1 to past 85% in long or saturated sessions, and fits ~3× more turns in the same context window. Powered by our open-source code-native 4B model. Drop-in for Claude Code, Cursor, Codex, OpenHands, and any BASE_URL agent.
Runssource: GitHubhomepagePythonApache-2.0commit c2ab50fc92fa
Python, Apache-2.0 licensed. The project labels itself: agentic coding, ai, ai agents, ai coding, ai infrastructure, developer tools, llm gateway and llms.
paritok-4b-v1 runs. An Argusic agent installed it in 2 minutes and hit 1 error and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
Python package installed in a venv, the proxy serves on port 8088, answers /health (200) and /stats (200) and /v1/messages/count_tokens (200), and all 316 tests in the test suite 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 96 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.
- Ollama not available (no root). Compression model cannot be pulled.
- Install time
- 2 minutes
- Cold machine to finish
- 8 minutes
- Errors hit and fixed
- 1 hit, 1 fixed with no human help
- How the result was proved
- pytest tests/ -v: 316 passed, 1 skipped; starlette TestClient: GET /health → 200, POST /v1/messages/count_tokens → 200, GET /stats → 200
- Model tokens used
- 52,336
- Exact commit tested
- c2ab50fc92fa
- 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
Watch the session
The whole run, from a clean machine, at the speed it actually happened. Nothing is cut and nothing is sped up.
A recording of attempt 2, run , the attempt that decided the verdict above. The terminal recording it was made from is the evidence, stored with a sha256 fingerprint; the video is a rendering of it. Nothing loads from the video host until you press play.
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
tokens used
52,336
Here's a summary of what I found and verified:
**Install**: Created a Python venv, installed `.[proxy,dev]` from the local repo. All dependencies …
**Tests**: Ran `pytest tests/ -v` — **316 passed, 1 skipped** (the skipped test `test_bfcl_recall_a…
**Proxy verification**: Used `starlette.testclient` to hit all key endpoints:
- `GET /health` → `200 {"status":"ok","version":"1.3.11"}`
- `POST /v1/messages/count_tokens` → `200 {"input_tokens": 14}`
- `GET /stats` → `200` with full stats payload (19 keys: uptime, token counts, etc.)
- `POST /v1/chat/completions` → `401` (correct — missing API key, proxy tries to forward upstream)
**Compression model**: The proxy requires Ollama running with `paritok/paritok-4b-v1` for local com…
**Verdict**: The repository works as-is for a brand-new user. All Python code installs, the proxy s…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 2 minutes.
- Recovered from all 1 error 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.
Of the 9 agentic-coding projects Argusic has installed and timed, paritok-4b-v1 was the 3rd fastest to reach a running state, and 8 of 9 reached one at all.
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 2 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
| tested by | attempt | outcome | score | cost USD | when |
|---|---|---|---|---|---|
| Argusic Runner | 2/3 | Runs | 100.00 | 0.04 | |
| Argusic Runner | 1/3 | Runs with mocks | 92.00 | 0.17 |
Topics (from GitHub)
agentic-codingaiai-agentsai-codingai-infrastructuredeveloper-toolsllm-gatewayllmstoken-optimization
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/paritok-4b-v1)Questions
- Does paritok-4b-v1 run?
- Yes. paritok-4b-v1 runs. Argusic installed and launched it on a clean machine in 2 minutes, hitting 1 error on the way, and recorded the session.
- How did Argusic test paritok-4b-v1?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit c2ab50fc92fa. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test paritok-4b-v1?
- The run that produced this verdict cost $0.04: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does paritok-4b-v1 take to install?
- 2 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 paritok-4b-v1 need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing paritok-4b-v1?
- 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 paritok-4b-v1 compare with the alternatives?
- Of the 9 agentic-coding projects Argusic has installed and timed, paritok-4b-v1 was the 3rd fastest to reach a running state, and 8 of 9 reached one at all.
- Where is the evidence for paritok-4b-v1?
- 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.