he commited on
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -77,12 +77,24 @@ library_name: transformers
|
|
| 77 |
|
| 78 |
M2.7 initiates a cycle of model self-evolution: during development, we let the model update its own memory, build dozens of complex skills for RL experiments, and improve its own learning process based on experiment results. An internal version of M2.7 autonomously optimized a programming scaffold over 100+ rounds — analyzing failure trajectories, modifying code, running evaluations, and deciding to keep or revert — achieving a **30% performance improvement**. On MLE Bench Lite (22 ML competitions), M2.7 achieved a **66.6% medal rate**, second only to Opus-4.6 and GPT-5.4.
|
| 79 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
## Professional Software Engineering
|
| 81 |
|
| 82 |
M2.7 delivers outstanding real-world programming capabilities spanning log analysis, bug troubleshooting, refactoring, code security, and machine learning. Beyond code generation, M2.7 demonstrates strong system-level reasoning — correlating monitoring metrics, conducting trace analysis, verifying root causes in databases, and making SRE-level decisions. Using M2.7, we have reduced live production incident recovery time to **under three minutes** on multiple occasions.
|
| 83 |
|
| 84 |
On SWE-Pro, M2.7 achieved **56.22%**, matching GPT-5.3-Codex, with even stronger performance on real-world engineering benchmarks: **SWE Multilingual (76.5)** and **Multi SWE Bench (52.7)**. On **VIBE-Pro (55.6%)**, M2.7 is nearly on par with Opus 4.6. On **Terminal Bench 2 (57.0%)** and **NL2Repo (39.8%)**, M2.7 demonstrates deep understanding of complex engineering systems. M2.7 also supports native **Agent Teams** for multi-agent collaboration with stable role identity and autonomous decision-making.
|
| 85 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
## Professional Work
|
| 87 |
|
| 88 |
M2.7 achieved an **ELO score of 1495** on GDPval-AA (highest among open-source models), surpassing GPT5.3. It handles Word, Excel, and PPT with high-fidelity multi-round editing, producing editable deliverables. On Toolathon, M2.7 reached **46.3%** accuracy (global top tier), and maintains **97% skill compliance** across 40+ complex skills on MM Claw. On the MM Claw end-to-end benchmark, M2.7 achieved **62.7%**, close to Sonnet 4.6.
|
|
|
|
| 77 |
|
| 78 |
M2.7 initiates a cycle of model self-evolution: during development, we let the model update its own memory, build dozens of complex skills for RL experiments, and improve its own learning process based on experiment results. An internal version of M2.7 autonomously optimized a programming scaffold over 100+ rounds — analyzing failure trajectories, modifying code, running evaluations, and deciding to keep or revert — achieving a **30% performance improvement**. On MLE Bench Lite (22 ML competitions), M2.7 achieved a **66.6% medal rate**, second only to Opus-4.6 and GPT-5.4.
|
| 79 |
|
| 80 |
+
<p align="center">
|
| 81 |
+
<img width="100%" src="figures/agent_harness.png">
|
| 82 |
+
</p>
|
| 83 |
+
|
| 84 |
+
<p align="center">
|
| 85 |
+
<img width="100%" src="figures/mle_bench.png">
|
| 86 |
+
</p>
|
| 87 |
+
|
| 88 |
## Professional Software Engineering
|
| 89 |
|
| 90 |
M2.7 delivers outstanding real-world programming capabilities spanning log analysis, bug troubleshooting, refactoring, code security, and machine learning. Beyond code generation, M2.7 demonstrates strong system-level reasoning — correlating monitoring metrics, conducting trace analysis, verifying root causes in databases, and making SRE-level decisions. Using M2.7, we have reduced live production incident recovery time to **under three minutes** on multiple occasions.
|
| 91 |
|
| 92 |
On SWE-Pro, M2.7 achieved **56.22%**, matching GPT-5.3-Codex, with even stronger performance on real-world engineering benchmarks: **SWE Multilingual (76.5)** and **Multi SWE Bench (52.7)**. On **VIBE-Pro (55.6%)**, M2.7 is nearly on par with Opus 4.6. On **Terminal Bench 2 (57.0%)** and **NL2Repo (39.8%)**, M2.7 demonstrates deep understanding of complex engineering systems. M2.7 also supports native **Agent Teams** for multi-agent collaboration with stable role identity and autonomous decision-making.
|
| 93 |
|
| 94 |
+
<p align="center">
|
| 95 |
+
<img width="100%" src="figures/agent_teams.gif">
|
| 96 |
+
</p>
|
| 97 |
+
|
| 98 |
## Professional Work
|
| 99 |
|
| 100 |
M2.7 achieved an **ELO score of 1495** on GDPval-AA (highest among open-source models), surpassing GPT5.3. It handles Word, Excel, and PPT with high-fidelity multi-round editing, producing editable deliverables. On Toolathon, M2.7 reached **46.3%** accuracy (global top tier), and maintains **97% skill compliance** across 40+ complex skills on MM Claw. On the MM Claw end-to-end benchmark, M2.7 achieved **62.7%**, close to Sonnet 4.6.
|