<div align="center">
<img src=assets/ornith_logo.png width="65%"/>
</div>
<div align="center">
Ornith Blog
</div>
Ornith-1.0
Aloha! 🌺 Ornith-1.0은 에이전트 기반 코딩 (agentic coding)을 위한 자기 개선형 (self-improving) 오픈 소스 모델입니다.
주요 특징:
- 최첨단 코딩 에이전트 (State-of-the-Art Coding Agents): 9B-Dense, 31B-Dense, 35B-MoE, 그리고 397B-MoE (Gemma 4 및 Qwen 3.5를 기반으로 사후 학습 (post-trained)됨) 모델로 제공되며, Terminal-Bench 2.1, SWE-Bench, NL2Repo, OpenClaw와 같은 코딩 벤치마크에서 유사한 크기의 오픈 소스 모델 중 최첨단 성능을 달성했습니다.
- 자기 개선형 학습 프레임워크 (Self-Improving Training Framework): Ornith-1.0은 강화학습 (RL)을 사용하여 솔루션 롤아웃 (solution rollouts)뿐만 아니라, 해당 롤아웃을 구동하는 스캐폴드 (scaffold)를 생성하는 법을 학습합니다. 스캐폴드와 그 결과로 나오는 솔루션을 공동으로 최적화함으로써, 모델은 더 나은 탐색 궤적 (search trajectories)을 발견하고 더 높은 품질의 솔루션을 생성합니다.
- 라이선스 (Licence): MIT 라이선스로 제공되며, 전 세계적으로 접근 가능하고 지역적 제한이 없습니다.
<img style="width: 100%; max-width: 900px;" src="https://raw.githubusercontent.com/deepreinforce-ai/Ornith-1/HEAD/assets/ornith_397b_eval.png" alt="Ornith 397B Benchmark Results" title="Ornith 397B Benchmark Results">
벤치마크 (Benchmarks)
각 모델은 크기에 적합한 베이스라인 (baselines)과 비교하여 평가됩니다. 세 모델 모두 동일한 하네스 (harnesses)와 디코딩 설정을 사용합니다 (표 아래의 노트를 참조하세요).
Ornith-1.0-9B
Ornith-1.0-9B
<div style="font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,sans-serif;width:100%;margin:0 auto;padding:16px 0">
<table style="width:100%;table-layout:fixed;border-collapse:collapse;font-size:13px">
<thead><tr>
<th style="width:28%;padding:10px 7px;text-align:left;font-weight:600;border-bottom:2px solid #FD8E5B;color:#FD8E5B"></th>
<th style="width:14.40%;padding:10px 7px;text-align:center;font-weight:700;border-bottom:2px solid #FD8E5B;color:#FD8E5B;font-size:14px;background:rgba(253, 142, 91, 0.12)">Ornith-1.0-9B</th>
<th style="width:14.40%;padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #FD8E5B;color:#FD8E5B;font-size:14px;">Qwen3.5-9B</th>
<th style="width:14.40%;padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #FD8E5B;color:#FD8E5B;font-size:14px;">Qwen3.5-35B</th>
<th style="width:14.40%;padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #FD8E5B;color:#FD8E5B;font-size:14px;">Gemma4-12B</th>
<th style="width:14.40%;padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #FD8E5B;color:#FD8E5B;font-size:14px;">Gemma4-31B</th>
</tr></thead>
<tbody>
<tr><td colspan="6" style="padding:8px 12px;font-weight:600;color:#FD8E5B;border-bottom:1px solid rgba(253, 142, 91, 0.2);background:rgba(253, 142, 91, 0.1)">Agentic Coding</td></tr>
</tbody>
</div>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">Terminal-Bench 2.1 <sub style="font-size: small;">(Terminus-2)</sub></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">43.1</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">21.3</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">41.4</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">21</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">42.1</td></tr>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">Terminal-Bench 2.1 <sub style="font-size: small;">(Claude Code)</sub></td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">40.6</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">18.9</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">38.9</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);">-</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);">-</td></tr>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">SWE-bench Verified</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">69.4</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">53.2</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">70</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">44.2</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">52</td></tr>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">SWE-bench Pro</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">42.9</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">31.3</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">44.6</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">27.6</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">35.7</td></tr>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">SWE-bench Multilingual</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">52</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">39.7</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">60.3</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">32.5</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">51.7</td></tr>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">NL2Repo</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">27.2</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">16.2</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">20.5</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">10.3</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">15.5</td></tr>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">Claw-eval Avg</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">63.1</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">53.2</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">65.4</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">32.5</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">48.5</td></tr>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">SWE Atlas - QnA</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">17.9</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">9.2</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">13.2</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);">-</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);">-</td></tr>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">SWE Atlas - RF</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">16.6</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">4.3</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">10.2</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">-</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">-</td></tr>
<tr><td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">SWE Atlas - TW</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);font-weight:600;color:#FD8E5B;background:rgba(253, 142, 91, 0.06)">15.3</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">4.4</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">9.8</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">-</td><td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">-</td></tr>
</tbody>
</table>
</div>
Ornith-1.0-35B