Prism-Shadow / GDPevo
A Benchmark for Evaluating Agent Self-Evolution on Real Business Tasks
Why this job: says: benchmark; topics: benchmark
What it does
Each experiment is run 3 times. Values in parentheses are standard deviations; all other values are means. Accuracy is reported as acc. rounds reports the average number of solver model-response rounds per attempt, and tool calls reports the average number of solver tool-call requests per attempt. Cost is reported in USD; lift and cost change are relative to base.
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit 11 days ago
- Stars
- 121
- +2 this week
- Licence
- Apache-2.0
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- None yet
- No releases
- Built with
- Python
- 91% of the code
Is it looked after?
- Contributors
- Not stated
- Last 90 days
- Issues
- 0% closed
- 2 still open
- Pull requests
- 46 merged
- 0 waiting
- Releases
- None
- Never published
- No contributing guide
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| suyoumo / ClawProBench | 829 | 2 months ago | Apache-2.0 | No downloads |
| alibaba / sec-code-bench | 138 | 28 days ago | Apache-2.0 | No downloads |
| sierra-research / tau2-bench | 2,419 | 5 days ago | MIT | No downloads |
More: languages, topics and project record
Languages
- Python 91%
- Shell 4%
- JavaScript 3%
- CSS 1%
Topics
- agent
- ai
- benchmark
- llm
- self-evolving
- self-learning
- skills
Project record
- JobEvals and observability
- StatusListed
- Commits17
- Forks5
- Watching1
- Stars in 28 days+55
- CreatedJun 5, 2026
- First seen by usNov 7, 2026
- CheckedNov 8, 2026