RonanMcGovern
commited on
Commit
•
39c4036
1
Parent(s):
d06f7b2
update for launch of mistral 7b function calling
Browse files
README.md
CHANGED
@@ -12,22 +12,25 @@ tags:
|
|
12 |
- functions
|
13 |
- function calling
|
14 |
- sharded
|
15 |
-
- gguf
|
16 |
-
- gptq
|
17 |
---
|
18 |
-
# Function Calling Llama 2 (version 2)
|
19 |
- Function calling Llama extends the hugging face Llama 2 models with function calling capabilities.
|
20 |
- The model responds with a structured json argument with the function name and arguments.
|
21 |
|
|
|
|
|
|
|
|
|
22 |
**Improvements with v2**
|
23 |
1. Shortened syntax: Only function descriptions are needed for inference and no added instruction is required.
|
24 |
2. Function descriptions are moved outside of the system prompt. This avoids the behaviour of function calling being affected by how the system prompt had been trained to influence the model.
|
25 |
|
26 |
Available models:
|
27 |
-
- Llama-7B-chat with function calling ([Base Model](https://huggingface.co/Trelis/Llama-2-7b-chat-hf-function-calling-v2)), ([PEFT Adapters](https://huggingface.co/Trelis/Llama-2-7b-chat-hf-function-calling-adapters-v2)), ([GGUF - files are in
|
28 |
-
-
|
29 |
-
-
|
30 |
-
-
|
|
|
31 |
|
32 |
## Performance and Tips
|
33 |
|
@@ -41,7 +44,7 @@ Check out this video overview of performance [here](https://www.loom.com/share/8
|
|
41 |
|
42 |
Llama-7B with function calling is licensed according to the Meta Community license.
|
43 |
|
44 |
-
Llama-13B, Code-llama-34b
|
45 |
- Commercial license purchase required per user.
|
46 |
- Licenses are not transferable to other users/entities.
|
47 |
|
|
|
12 |
- functions
|
13 |
- function calling
|
14 |
- sharded
|
|
|
|
|
15 |
---
|
16 |
+
# Function Calling Llama 2 + Mistral Models (version 2)
|
17 |
- Function calling Llama extends the hugging face Llama 2 models with function calling capabilities.
|
18 |
- The model responds with a structured json argument with the function name and arguments.
|
19 |
|
20 |
+
**Recent Updates**
|
21 |
+
- October 11th 2023 -> added Mistral 7B with function calling
|
22 |
+
- October 11th 2023 -> new models pushed, trained on an improved underlying dataset
|
23 |
+
|
24 |
**Improvements with v2**
|
25 |
1. Shortened syntax: Only function descriptions are needed for inference and no added instruction is required.
|
26 |
2. Function descriptions are moved outside of the system prompt. This avoids the behaviour of function calling being affected by how the system prompt had been trained to influence the model.
|
27 |
|
28 |
Available models:
|
29 |
+
- Llama-7B-chat with function calling ([Base Model](https://huggingface.co/Trelis/Llama-2-7b-chat-hf-function-calling-v2)), ([PEFT Adapters](https://huggingface.co/Trelis/Llama-2-7b-chat-hf-function-calling-adapters-v2)), ([GGUF - files are in the main branch of the base model]) - Free
|
30 |
+
- Mistral-7B-Instruct-v0.1 with function calling ([Base Model](https://huggingface.co/Trelis/Mistral-7B-Instruct-v0.1-function-calling-v2)), ([PEFT Adapters](https://huggingface.co/Trelis/Mistral-7B-Instruct-v0.1-function-calling-adapters-v2)) - Paid, [purchase here](https://buy.stripe.com/cN2cNybSdgyncV25kQ)
|
31 |
+
- Llama-13B-chat with function calling ([Base Model](https://huggingface.co/Trelis/Llama-2-13b-chat-hf-function-calling-v2)), ([PEFT Adapters](https://huggingface.co/Trelis/Llama-2-13b-chat-hf-function-calling-adapters-v2)) - Paid, [purchase here](https://buy.stripe.com/9AQ7te3lHdmbdZ68wz)
|
32 |
+
- CodeLlama-34B-Instruct with function calling ([Base Model](https://huggingface.co/Trelis/CodeLlama-34b-Instruct-hf-function-calling-v2)), ([PEFT Adapters](https://huggingface.co/Trelis/CodeLlama-34b-Instruct-hf-function-calling-adapters-v2)) - Paid, [purchase here](https://buy.stripe.com/cN27teg8t2Hx5sA8wM)
|
33 |
+
- Llama-70B-chat with function calling ([Base Model](https://huggingface.co/Trelis/Llama-2-70b-chat-hf-function-calling-v2)), ([PEFT Adapters](https://huggingface.co/Trelis/Llama-2-70b-chat-hf-function-calling-adapters-v2)) - Paid, [purchase here](https://buy.stripe.com/8wMdRC1dzci75sA4gy)
|
34 |
|
35 |
## Performance and Tips
|
36 |
|
|
|
44 |
|
45 |
Llama-7B with function calling is licensed according to the Meta Community license.
|
46 |
|
47 |
+
Mistral-7B, Llama-13B, Code-llama-34b, Llama-70B and Falcon-180B with function calling require the purchase of access.
|
48 |
- Commercial license purchase required per user.
|
49 |
- Licenses are not transferable to other users/entities.
|
50 |
|