AgentKernelArena provides an end-to-end siloed-benchmarking environment where different LLM-powered agents—such as Cursor Agent, Claude Code, Codex, SWE-agent, and GEAK—can be evaluated side-by-side on the same GPU kernel tasks, using objective and reproducible metrics.
active 2026-01-16 → 2026-08-19 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 216 days from 2026-01-16 to 2026-08-19. Pushes: 93 total, peak 9 in a day. Pull requests: 5 total, peak 1 in a day. Issues: 3 total, peak 3 in a day. Comments: 7 total, peak 2 in a day. Stars: 7 total, peak 2 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| irvineoy | 43 | 32 | 3 | 5 |
| iraj465 | 20 | 20 | 0 | 0 |
| yueliu14 | 11 | 11 | 0 | 0 |
| mattwill-amd | 8 | 8 | 0 | 0 |
| johayang-amd | 7 | 7 | 0 | 0 |
| Puyuan-Yang | 6 | 6 | 0 | 0 |
| mehdi-saeedi | 4 | 3 | 1 | 0 |
| sharareh-y | 4 | 1 | 1 | 2 |
| sinarafati-amd | 3 | 3 | 0 | 0 |
| sdubagun-amd | 2 | 2 | 0 | 0 |
| amd-vivekag | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Pull request#35mehdi-saeedi2026-05-07 00:32
- Pull request#34sharareh-y2026-05-01 00:08
- Pull request#29irvineoy2026-04-13 23:59
- Issue comment#27irvineoy2026-03-31 00:34[FIX] chaned loaded kernel from .cu to .hip
- Issue comment#24irvineoy2026-03-12 17:36Tasks Improvement: Enhanced Benchmarking and Task Organization
- Issue comment#23irvineoy2026-03-04 17:48Sharareh/unified evaluator
- Pull request#21irvineoy2026-02-25 15:06
- Issue comment#20sharareh-y2026-02-24 19:23Improve OpenEvolve launcher setup, run performance eval for speedup, and standardize task result output
- Pull request#17irvineoy2026-02-19 19:14
- Issue#11irvineoy2026-01-26 16:26Benchmarking State-of-the-Art Agents for Technical Reporting
- Issue#5irvineoy2026-01-26 15:59Expand Pytorch2HIP Task Set to 100+ Tasks
- Issue#4irvineoy2026-01-26 15:59Expand CUDA2HIP Task Set to 100+ Tasks
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 7 stars here means stars gained during the window, not the repo's star count.