Update README.md
Browse files
README.md
CHANGED
|
@@ -19,11 +19,19 @@ MODEL=accounts/fireworks/models/firefunction-v1
|
|
| 19 |
|
| 20 |
## Intended Use and Limitations
|
| 21 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
### Primary Use
|
| 23 |
Although the model was trained on a variety of tasks, it performs best on:
|
| 24 |
* single-turn request routing to a function picked from a pool of up to 20 function specs.
|
| 25 |
* structured information extraction.
|
| 26 |
|
|
|
|
|
|
|
| 27 |
### Out-of-Scope Use
|
| 28 |
The model was not optimized for the following use cases:
|
| 29 |
* general multi-turn chat,
|
|
|
|
| 19 |
|
| 20 |
## Intended Use and Limitations
|
| 21 |
|
| 22 |
+
### Key Highlights
|
| 23 |
+
⭐️ Near GPT-4 level quality for real-world use cases of structured information generation and routing decision-making
|
| 24 |
+
💨 Blazing fast speed. Inference speeds are roughly 4x that of GPT-4 when using FireFunction hosted on the Fireworks platform
|
| 25 |
+
🔄 Support for "any" paramter in tool_choice. Firefunction is the only model that we're aware that supports an option for the model to always choose a function - particularly helpful for routing use cases
|
| 26 |
+
|
| 27 |
+
|
| 28 |
### Primary Use
|
| 29 |
Although the model was trained on a variety of tasks, it performs best on:
|
| 30 |
* single-turn request routing to a function picked from a pool of up to 20 function specs.
|
| 31 |
* structured information extraction.
|
| 32 |
|
| 33 |
+
|
| 34 |
+
|
| 35 |
### Out-of-Scope Use
|
| 36 |
The model was not optimized for the following use cases:
|
| 37 |
* general multi-turn chat,
|