Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,55 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: meta-llama/Meta-Llama-3.1-70B-Instruct
|
| 4 |
+
tags:
|
| 5 |
+
- function
|
| 6 |
+
- function-calling
|
| 7 |
+
- tool-using
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
## Empower Functions Model v1.1
|
| 11 |
+
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
[https://github.com/empower-ai/empower-functions](https://github.com/empower-ai/empower-functions)
|
| 15 |
+
|
| 16 |
+
Empower Functions is a family of LLMs(large language models) that offer GPT-4 level capabilities for real-world "tool using" use cases, with full compatibility support to serve as a drop-in replacement.
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
## Key Features
|
| 20 |
+
* Automatic tool using, able to decide when to use tools and when to converse, optimized for long conversations
|
| 21 |
+
* Parallel call, supports calling one function multiple times, multiple functions, or a combination of both
|
| 22 |
+
* Sequential calling, supports calling multiple functions sequentially to fulfill the user request
|
| 23 |
+
* Streaming
|
| 24 |
+
|
| 25 |
+
## Family of Models
|
| 26 |
+
|
| 27 |
+
| Model | Specs | Links | Notes |
|
| 28 |
+
| ------------------------------ | ------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------- |
|
| 29 |
+
| llama3-empower-functions-small | 128k context, based on [Llama3.1 8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) | [model](https://huggingface.co/empower-dev/llama3-empower-functions-small-v1.1), [gguf](https://huggingface.co/empower-dev/llama3-empower-functions-small-gguf-v1.1) | Most cost-effective, locally runnable | | Balance in accuracy and cost |
|
| 30 |
+
| llama3-empower-functions-large | 128k context, based on [Llama3.1 70B](https://huggingface.co/meta-llama/Meta-Llama-3.1-70B) | [model](https://huggingface.co/empower-dev/llama3-empower-functions-large-v1.1) | Best accuracy |
|
| 31 |
+
|
| 32 |
+
### Hardware Requirement
|
| 33 |
+
|
| 34 |
+
We have tested and the family of models in following setup:
|
| 35 |
+
|
| 36 |
+
- empower-functions-small: fp16 on 1xA100 40G, GGUF and 4bit GGUF on Macbook M2 Pro with 32G RAM, in minimal the 4bit GGUF version requires 7.56G RAM.
|
| 37 |
+
- empower-functions-medium: fp16 on 2xA100 80G
|
| 38 |
+
- empower-functions-large: fp16 on 4xA100 80G
|
| 39 |
+
|
| 40 |
+
## Usage
|
| 41 |
+
|
| 42 |
+
There are three ways to use the empower-functions model. You can either directly [prompt the raw model](https://github.com/empower-ai/empower-functions?tab=readme-ov-file#prompt-raw-model), run it [locally](https://github.com/empower-ai/empower-functions?tab=readme-ov-file#running-locally) through llama-cpp-python, or use our [hosted API](https://github.com/empower-ai/empower-functions?tab=readme-ov-file#using-empower-api)
|
| 43 |
+
|
| 44 |
+
## Evaluation
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
v1.1 is the newer version trained based on meta llama3.1 with the newly updated dataset, it has achieved state-of-the-art performance on the Berkeley Function Calling leaderboard:
|
| 48 |
+
|
| 49 |
+
|
| 50 |
+
(captured on Sep 10, 2024)
|
| 51 |
+
|
| 52 |
+
## Demo App
|
| 53 |
+
Check our healthcare appointment booking [demo](https://app.empower.dev/chat-demo)
|
| 54 |
+
|
| 55 |
+
Want to customize the model? Please contact us at [[email protected]](mailto:[email protected])
|