AI Accelerator Benchmark focuses on evaluating AI Accelerators from a practical production perspective, including the ease of use and versatility of software and hardware.
active 2023-08-16 → 2025-12-10 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 848 days from 2023-08-16 to 2025-12-10. Pushes: 249 total, peak 9 in a day. Pull requests: 146 total, peak 7 in a day. Issues: 52 total, peak 8 in a day. Comments: 143 total, peak 10 in a day. Stars: 168 total, peak 4 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 |
|---|---|---|---|---|
| suisiyuan | 194 | 101 | 60 | 25 |
| JianzheXiao | 93 | 53 | 21 | 10 |
| YJessicaGao | 67 | 25 | 17 | 10 |
| HanTengfei99 | 42 | 22 | 11 | 5 |
| Minghui-BD | 32 | 20 | 0 | 6 |
| CLAassistant | 31 | 0 | 0 | 31 |
| ZhaiFeiyue | 26 | 0 | 0 | 13 |
| kevinsouthByteDance | 22 | 19 | 1 | 1 |
| Yi-sir | 13 | 0 | 1 | 7 |
| HunterHantao | 12 | 9 | 2 | 1 |
| terU3760 | 11 | 0 | 0 | 5 |
| keg0704 | 9 | 0 | 3 | 3 |
| wanto-rest | 6 | 0 | 0 | 5 |
| incomingflyingbrick | 5 | 0 | 0 | 2 |
| eric-yq | 5 | 0 | 0 | 1 |
| MikuGhoul | 4 | 0 | 4 | 0 |
| yizhou-xu | 4 | 0 | 0 | 2 |
| gonggqing | 4 | 0 | 4 | 0 |
| DeepTecher | 4 | 0 | 0 | 1 |
| wei1tang | 4 | 0 | 1 | 3 |
Recent activity
Latest issues, pull requests and releases
- Issue#161zlk03062025-12-09 01:42关于byte_micro_perf中quant_matmul算子的配置问题
- Issue comment#146Riverstrider2025-12-08 13:47关于模型和算子的测试精度标准
- Issue#156zhemutsu2025-12-08 12:28llmperf精度测试提到跟A100比较，但是没有pass准则，需确认
- Issue#153zlk03062025-12-08 11:40关于byte_infer_perf中的模型修改问题
- Issue#152zlk03062025-12-08 09:02关于byte_infer_perf中的attention_mask和causal（因果掩码）
- Issue#150suisiyuan2025-12-08 07:15关于byte_micro_perf中flash attention session的varlen问题
- Issue comment#150suisiyuan2025-12-08 07:11关于byte_micro_perf中flash attention session的varlen问题
- Issue comment#144HunterHantao2025-12-08 07:11内存使用和碎片问题
- Issue#149zlk03062025-12-08 06:56byte_train_perf中的模型配置问题
- Issue#148xqdan2025-12-08 05:27llm推理模型list
- Issue#144xqdan2025-12-08 05:17内存使用和碎片问题
- Issue comment#143suisiyuan2025-12-08 04:08tensorflow/pytorch版本问题
- Issue#142xqdan2025-12-08 03:46flash_attention测试用例存在QKV量化位宽支持不一致
- Pull request#141yupengzh-intel2025-08-27 08:21
- Pull request#140HunterHantao2025-08-18 13:08
- Issue comment#140CLAassistant2025-08-18 12:56Training v0.1
- Pull request#140HunterHantao2025-08-18 12:56
- Issue comment#138caizhi-mt2025-06-27 08:15修复在跑算子测试时，其余卡查询0卡的可用内存问题
- Pull request#138sunyanguomt2025-06-27 07:57
- Issue#136wangrt12025-06-27 03:52【byte_micro_perf】为什么用 profiler 替换 torch.cuda.Event 统计时间呢
- Issue comment#136wangrt12025-06-27 03:52【byte_micro_perf】为什么用 profiler 替换 torch.cuda.Event 统计时间呢
- Issue comment#136wangrt12025-06-18 13:20【byte_micro_perf】为什么用 profiler 替换 torch.cuda.Event 统计时间呢
- Issue comment#135CLAassistant2025-05-24 11:26fix(flash_mla): fix a shape bug in flash_mla test
- Pull request#134brianlcy1232025-04-01 05:07
- Issue comment#134CLAassistant2025-04-01 05:07Cyliu/klx
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 168 stars here means stars gained during the window, not the repo's star count.