Add model card for Solar Open 100B

README.md

@@ -1,5 +1,60 @@
----
-license: other
-license_name: solar-apache-license
-license_link: LICENSE
----
+---
+language:
+- en
+- ko
+license: other
+license_name: solar-apache-2.0
+tags:
+- upstage
+- solar
+- moe
+- 100b
+- llm
+---
+
+# **Solar Open**
+
+**Solar Open** is Upstage's flagship large language model with **102B parameters**, now available under the **Solar-Apache License 2.0** (see LICENSE file). It is a **Mixture-of-Experts (MoE)** model designed to empower the open-source community with enterprise-grade reasoning, instruction-following, and agentic capabilities.
+
+## Highlights
+* **MoE Architecture (102B / 12B):** Built on a Mixture-of-Experts architecture with **102B total / 12B active parameters**. This design delivers the knowledge depth of a massive model with the inference speed and cost-efficiency of a much smaller model.
+* **Agentic Specialist with Parallel Tool Calling:** Engineered to handle complex agentic workflows. It supports **Parallel Tool Calling**, allowing the model to generate multiple function calls in a single turn to execute tasks efficiently.
+* **Massive Training Scale:** Pre-trained on **19.7 trillion tokens**, ensuring broad knowledge coverage and robust reasoning capabilities across various domains.
+
+## Model Overview
+* **Model Name:** Solar Open 100B
+* **Hugging Face ID:** Upstage/Solar-Open-100B
+* **Architecture:** Mixture-of-Experts (MoE)
+    * **Total Parameters:** 102.6B
+    * **Active Parameters:** 12B (per token)
+    * **Experts:** 129 Experts (top 8 among 128 Routed + 1 Shared)
+* **Pre-training Tokens:** 19.7 Trillion
+* **Context Length:** 128k
+* **Training Hardware:** NVIDIA B200 GPUs
+* **License:** **Solar-Apache License 2.0** (See `LICENSE` file)
+
+## Performance
+*Detailed benchmarks and performance metrics will be updated upon the official release on December 31, 2025.*
+
+## Quickstart
+*Python code snippets and usage examples will be available upon the official release on December 31, 2025.*
+
+## Agentic Use & Parallel Tool Calling
+Solar Open excels at **Parallel Tool Calling**, enabling the model to request multiple actions simultaneously within a single turn. This reduces latency and improves the efficiency of AI agents.
+
+## Public API Access
+The official API service for Solar Open is scheduled to launch publicly on **January 1, 2026**.
+* **Access:** Upstage Console (Available starting Jan 1, 2026)
+* **Documentation:** [**Upstage Console Docs**](https://console.upstage.ai/docs/getting-started)
+
+## Citation
+If you use Solar Open in your research, please cite:
+```bibtex
+@misc{solar-open-2025,
+    title={Solar Open: Scaling Upstage's LLM Capabilities with MoE},
+    author={Upstage AI},
+    year={2025},
+    url={https://huggingface.co/Upstage/Solar-Open-100B}
+}
+```
+