|
--- |
|
license: other |
|
pipeline_tag: text-generation |
|
tags: |
|
- cortex.cpp |
|
--- |
|
|
|
## Overview |
|
|
|
Athene-V2-Chat-72B is an open-weight LLM that competes on par with GPT-4o across various benchmarks. It is currently ranked as the best open model on Chatbot Arena, where it outperforms GPT-4o-0513 (the highest-ranked GPT-4o model on Arena) in hard and math categories. It also matches GPT-4o-0513 in coding, instruction following, longer queries, and multi-turn conversations. |
|
|
|
Trained through RLHF with Qwen-2.5-72B-Instruct as the base model, Athene-V2-Chat-72B excels in chat, math, and coding. Additionally, its sister model, Athene-V2-Agent-72B, surpasses GPT-4o in complex function calling and agentic applications, further extending its capabilities. |
|
## Variants |
|
|
|
| No | Variant | Cortex CLI command | |
|
| --- | --- | --- | |
|
| 1 | [Athene-72b](https://huggingface.co/cortexso/athene/tree/72b) | `cortex run athene:72b` | |
|
|
|
## Use it with Jan (UI) |
|
|
|
1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart) |
|
2. Use in Jan model Hub: |
|
```bash |
|
cortexhub/athene |
|
``` |
|
|
|
## Use it with Cortex (CLI) |
|
|
|
1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart) |
|
2. Run the model with command: |
|
```bash |
|
cortex run athene |
|
``` |
|
|
|
## Credits |
|
|
|
- **Author:** Nexusflow |
|
- **Converter:** [Homebrew](https://homebrew.ltd/) |
|
- **Original License:** [Licence](https://huggingface.co/Nexusflow/Athene-V2-Chat/blob/main/Nexusflow_Research_License_.pdf) |
|
- **Papers:** [Athene V2 Blog](https://nexusflow.ai/blogs/athene-v2) |