Update README.md from tmp
Browse files
README.md
CHANGED
|
@@ -4,7 +4,7 @@ language:
|
|
| 4 |
- ko
|
| 5 |
library_name: transformers
|
| 6 |
license: other
|
| 7 |
-
license_name: solar-
|
| 8 |
pipeline_tag: text-generation
|
| 9 |
tags:
|
| 10 |
- upstage
|
|
@@ -21,7 +21,7 @@ arxiv: 2601.07022
|
|
| 21 |
|
| 22 |
# **Solar Open**
|
| 23 |
|
| 24 |
-
**Solar Open** is Upstage's flagship **102B-parameter** large language model, trained **entirely from scratch** and released under the **Solar
|
| 25 |
|
| 26 |
[**Technical Report**](https://huggingface.co/papers/2601.07022) | [**Project Page**](https://upstage.ai)
|
| 27 |
|
|
@@ -41,7 +41,7 @@ arxiv: 2601.07022
|
|
| 41 |
* **Pre-training Tokens:** 19.7 Trillion
|
| 42 |
* **Context Length:** 128k
|
| 43 |
* **Training Hardware:** NVIDIA B200 GPUs
|
| 44 |
-
* **License:** **Solar
|
| 45 |
* **Hardware Requirements:**
|
| 46 |
* **Minimum:** 4x NVIDIA A100 (80GB)
|
| 47 |
|
|
@@ -52,7 +52,7 @@ This repository contains both model weights and code,
|
|
| 52 |
which are licensed under different terms:
|
| 53 |
|
| 54 |
1. MODEL WEIGHTS (*.safetensors)
|
| 55 |
-
Licensed under **Solar
|
| 56 |
See: https://huggingface.co/upstage/Solar-Open-100B/blob/main/LICENSE
|
| 57 |
|
| 58 |
2. CODE (*.py, *.json, *.jinja files)
|
|
@@ -215,7 +215,7 @@ vllm serve upstage/Solar-Open-100B \
|
|
| 215 |
|
| 216 |
## Public API Access
|
| 217 |
|
| 218 |
-
The official API service for Solar Open is scheduled to launch publicly
|
| 219 |
|
| 220 |
* **Access:** Upstage Console (TBA)
|
| 221 |
* **Documentation:** Upstage Console (TBA)
|
|
|
|
| 4 |
- ko
|
| 5 |
library_name: transformers
|
| 6 |
license: other
|
| 7 |
+
license_name: upstage-solar-license
|
| 8 |
pipeline_tag: text-generation
|
| 9 |
tags:
|
| 10 |
- upstage
|
|
|
|
| 21 |
|
| 22 |
# **Solar Open**
|
| 23 |
|
| 24 |
+
**Solar Open** is Upstage's flagship **102B-parameter** large language model, trained **entirely from scratch** and released under the **Upstage Solar License** (see [LICENSE](#license) for details). As a **Mixture-of-Experts (MoE)** architecture, it delivers enterprise-grade performance in reasoning, instruction-following, and agentic capabilities—all while prioritizing transparency and customization for the open-source community.
|
| 25 |
|
| 26 |
[**Technical Report**](https://huggingface.co/papers/2601.07022) | [**Project Page**](https://upstage.ai)
|
| 27 |
|
|
|
|
| 41 |
* **Pre-training Tokens:** 19.7 Trillion
|
| 42 |
* **Context Length:** 128k
|
| 43 |
* **Training Hardware:** NVIDIA B200 GPUs
|
| 44 |
+
* **License:** **Upstage Solar License** (See [LICENSE](#license))
|
| 45 |
* **Hardware Requirements:**
|
| 46 |
* **Minimum:** 4x NVIDIA A100 (80GB)
|
| 47 |
|
|
|
|
| 52 |
which are licensed under different terms:
|
| 53 |
|
| 54 |
1. MODEL WEIGHTS (*.safetensors)
|
| 55 |
+
Licensed under **Upstage Solar License**
|
| 56 |
See: https://huggingface.co/upstage/Solar-Open-100B/blob/main/LICENSE
|
| 57 |
|
| 58 |
2. CODE (*.py, *.json, *.jinja files)
|
|
|
|
| 215 |
|
| 216 |
## Public API Access
|
| 217 |
|
| 218 |
+
The official API service for Solar Open is scheduled to launch publicly in **January**.
|
| 219 |
|
| 220 |
* **Access:** Upstage Console (TBA)
|
| 221 |
* **Documentation:** Upstage Console (TBA)
|