我们引入了一个专为代码生成任务设计的新模型。它在 HumanEval 基础数据集上的测试准确性超过了 GPT-4 Turbo（2024 年 4 月）和 GPT-4o。
active 2024-05-14 → 2026-03-26 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 682 days from 2024-05-14 to 2026-03-26. Pushes: 18 total, peak 5 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 14 total, peak 3 in a day. Comments: 19 total, peak 5 in a day. Stars: 610 total, peak 135 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 |
|---|---|---|---|---|
| bin123apple | 31 | 18 | 0 | 13 |
| xxyp | 3 | 0 | 0 | 1 |
| wikeeyang | 3 | 0 | 0 | 1 |
| KevinH48264 | 3 | 0 | 0 | 2 |
| Wawrzyn1985 | 3 | 0 | 0 | 0 |
| ethanc8 | 2 | 0 | 0 | 1 |
| zhoucz97 | 2 | 0 | 0 | 1 |
| jonathanpv | 2 | 0 | 0 | 0 |
| Network-Sec | 1 | 0 | 0 | 0 |
| chmod777john | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#8Wawrzyn19852024-12-25 20:38Dupek
- Issue#8Wawrzyn19852024-12-25 20:37Dupek
- Issue comment#9bin123apple2024-08-11 01:48Impossible to install
- Issue comment#10bin123apple2024-08-11 01:46Community Group
- Issue#13zhoucz972024-07-28 14:52Some problems about two-stage data synthesis
- Issue comment#13zhoucz972024-07-28 14:51Some problems about two-stage data synthesis
- Issue comment#13bin123apple2024-07-26 19:13Some problems about two-stage data synthesis
- Issue comment#11KevinH482642024-06-06 05:18What information is left out during inference from the output of model.generate()?
- Issue#7jonathanpv2024-06-06 04:29Support datagen on any arbirtrary github repo
- Issue comment#11bin123apple2024-06-05 23:43What information is left out during inference from the output of model.generate()?
- Issue comment#11bin123apple2024-06-03 04:08What information is left out during inference from the output of model.generate()?
- Issue comment#11KevinH482642024-06-03 03:59What information is left out during inference from the output of model.generate()?
- Issue comment#9bin123apple2024-06-03 01:00Impossible to install
- Issue comment#11bin123apple2024-06-03 00:46What information is left out during inference from the output of model.generate()?
- Issue comment#10bin123apple2024-06-03 00:29Community Group
- Issue#11KevinH482642024-06-02 20:00What information is left out during inference from the output of model.generate()?
- Issue#10chmod777john2024-06-02 00:23Community Group
- Issue comment#9bin123apple2024-06-01 22:39Impossible to install
- Issue#9Network-Sec2024-06-01 22:05Impossible to install
- Issue#8Wawrzyn19852024-06-01 19:44Grzesiek
- Issue comment#7bin123apple2024-05-31 23:50Support datagen on any arbirtrary github repo
- Issue#7jonathanpv2024-05-31 22:54Support datagen on any arbirtrary github repo
- Issue comment#6bin123apple2024-05-31 17:46AEIV-Instruct dataset
- Issue comment#6ethanc82024-05-31 16:07AEIV-Instruct dataset
- Issue#6ethanc82024-05-31 16:05AEIV-Instruct dataset
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 610 stars here means stars gained during the window, not the repo's star count.