MiroThinker
MiroThinker is a deep research agent optimized for complex research and prediction tasks. Our latest models, MiroThinker-1.7, achieves 74.0 and 75.3 on the BrowseComp and BrowseComp Zh, respectively.
Runs with mockssource: GitHubhomepagePythonApache-2.0commit 1c4253f6774b
Python, Apache-2.0 licensed. The project labels itself: agent, agent framework, browsecomp, deep research, futurex, gaia, hle and research agent.
MiroThinker runs, with stand-ins for the services it depends on. An Argusic agent installed it in 5 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 cleanly, the agent pipeline executes a real MCP tool through a local mock LLM and completes with status success and boxed answer 'answer-is-42', 39 lobehub parser tests pass, and the trace dashboard serves HTTP 200.
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
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.
- main.py crashed: StdioServerParameters env.E2B_API_KEY must be a string when tool-python is configured and .env is absent12 minutes
- My initial test hook patch had an indentation error, then missing stop-path (Unsupported finish reason: stop)8 minutes
- uv not installed in container; system Python is PEP 668 externally-managed4 minutes
- Install time
- 5 minutes
- Cold machine to finish
- 38 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- How the result was proved
- uv run python main.py llm=qwen-3 agent=single_agent_mock llm.base_url=http://127.0.0.1:8000/v1 agent.main_agent.max_turns=1 -> Tool Call Success, Boxed answer found on attempt 1, task_execution_finished ... status: success; pytest unit_test.py test_tool_parser.py -> 39 passed; curl -s -o /dev/null -w %{http_code} http://127.0.0.1:5001/ -> 200
- Model tokens used
- 3,923,243
- Exact commit tested
- 1c4253f6774b
- 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.
- `apps/visualize-trace`: `run.py -p 5001` → `curl /` returned **HTTP 200** with the dashboard HTML…
**Not claimed**
- Real Serper/Jina/E2B/Anthropic/Argusic calls (no real credentials; mocks stood up for Argusic-com…
- The agent's `tool-python` (real E2B) path — schema-correct mock only for that capability.
**Timing**
- ~11:07 → 11:44 UTC ≈ **37 minutes** of work.
**Final state**
- `uv sync` succeeds; full agent pipeline runs end-to-end against the local mock and completes with…
**Verified how**
- `uv run python main.py llm=qwen-3 agent=single_agent_mock llm.base_url=http://127.0.0.1:8000/v1 a…
- `/tmp/lobe-venv/bin/pytest unit_test.py test_tool_parser.py` → `39 passed in 0.71s`.
- `curl -s -o /dev/null -w '%{http_code}' http://127.0.0.1:5001/` → `200`.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 5 minutes, faster than the median of the 137 comparable projects Argusic has measured.
- Recovered from all 3 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.
Of the 138 agent projects Argusic has installed and timed, MiroThinker was the 55th fastest to reach a running state, and 92 of 138 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 | 2/3 | Runs with mocks | 92.00 | 0.81 | |
| Argusic Runner | 1/3 | Runs with mocks | 92.00 | 0.10 |
Topics (from GitHub)
agentagent-frameworkbrowsecompdeep-researchfuturexgaiahleresearch-agentsearch-agentxbench
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/mirothinker)Questions
- Does MiroThinker run?
- MiroThinker runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 5 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test MiroThinker?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 1c4253f6774b. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test MiroThinker?
- The run that produced this verdict cost $0.81: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does MiroThinker take to install?
- 5 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 MiroThinker need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing MiroThinker?
- 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.
- How does MiroThinker compare with the alternatives?
- Of the 138 agent projects Argusic has installed and timed, MiroThinker was the 55th fastest to reach a running state, and 92 of 138 reached one at all.
- Where is the evidence for MiroThinker?
- 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.