Skip to content

Commit 402c415

Browse files
mudlermudler
committed
feat(gallery): add llama3, hermes, phi-3, and others
Also adds embeddings and llava models Signed-off-by: Ettore Di Giacinto <[email protected]>
1 parent d344daf commit 402c415

11 files changed

+804
-229
lines changed

‎gallery/codellama.yaml

Lines changed: 21 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,21 @@
1+
name: "codellama"
2+
license: llama2
3+
4+
description: |
5+
Code Llama is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 34 billion parameters. This model is designed for general code synthesis and understanding.
6+
7+
urls:
8+
- https://huggingface.co/TheBloke/CodeLlama-7B-GGUF
9+
- https://huggingface.co/meta-llama/CodeLlama-7b-hf
10+
11+
tags:
12+
- llm
13+
- gguf
14+
- gpu
15+
- cpu
16+
17+
config_file: |
18+
backend: llama-cpp
19+
context_size: 4096
20+
f16: true
21+
mmap: true

‎gallery/dreamshaper.yaml

Lines changed: 26 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,26 @@
1+
name: "dreamshaper"
2+
icon: https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/ggO2sBDJ8Bhc6w-zwTx5j.png
3+
license: other
4+
5+
description: |
6+
A text-to-image model that uses Stable Diffusion 1.5 to generate images from text prompts. This model is DreamShaper model by Lykon.
7+
8+
urls:
9+
- https://civitai.com/models/4384/dreamshaper
10+
11+
tags:
12+
- text-to-image
13+
- stablediffusion
14+
- sd-1.5
15+
- gpu
16+
17+
config_file: |
18+
backend: diffusers
19+
step: 25
20+
f16: true
21+
22+
diffusers:
23+
pipeline_type: StableDiffusionPipeline
24+
cuda: true
25+
enable_parameters: "negative_prompt,num_inference_steps"
26+
scheduler_type: "k_dpmpp_2m"

‎gallery/hermes-2-pro-mistral.yaml

Lines changed: 81 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,81 @@
1+
name: "hermes-2-pro-mistral"
2+
icon: https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/ggO2sBDJ8Bhc6w-zwTx5j.png
3+
license: apache-2.0
4+
5+
description: |
6+
Hermes 2 Pro is an upgraded, retrained version of Nous Hermes 2, consisting of an updated and cleaned version of the OpenHermes 2.5 Dataset, as well as a newly introduced Function Calling and JSON Mode dataset developed in-house.
7+
8+
This new version of Hermes maintains its excellent general task and conversation capabilities - but also excels at Function Calling, JSON Structured Outputs, and has improved on several other metrics as well, scoring a 90% on our function calling evaluation built in partnership with Fireworks.AI, and an 81% on our structured JSON Output evaluation.
9+
10+
Hermes Pro takes advantage of a special system prompt and multi-turn function calling structure with a new chatml role in order to make function calling reliable and easy to parse. Learn more about prompting below.
11+
12+
This work was a collaboration between Nous Research, @interstellarninja, and Fireworks.AI
13+
14+
Learn more about the function calling on our github repo here: https://github.com/NousResearch/Hermes-Function-Calling/tree/main
15+
16+
urls:
17+
- https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B-GGUF
18+
19+
tags:
20+
- llm
21+
- gguf
22+
- gpu
23+
- cpu
24+
25+
config_file: |
26+
mmap: true
27+
parameters:
28+
model: Hermes-2-Pro-Mistral-7B.Q6_K.gguf
29+
30+
template:
31+
chat_message: |
32+
<|im_start|>{{if eq .RoleName "assistant"}}assistant{{else if eq .RoleName "system"}}system{{else if eq .RoleName "tool"}}tool{{else if eq .RoleName "user"}}user{{end}}
33+
{{- if .FunctionCall }}
34+
<tool_call>
35+
{{- else if eq .RoleName "tool" }}
36+
<tool_response>
37+
{{- end }}
38+
{{- if .Content}}
39+
{{.Content }}
40+
{{- end }}
41+
{{- if .FunctionCall}}
42+
{{toJson .FunctionCall}}
43+
{{- end }}
44+
{{- if .FunctionCall }}
45+
</tool_call>
46+
{{- else if eq .RoleName "tool" }}
47+
</tool_response>
48+
{{- end }}
49+
<|im_end|>
50+
# https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B-GGUF#prompt-format-for-function-calling
51+
function: |
52+
<|im_start|>system
53+
You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions. Here are the available tools:
54+
<tools>
55+
{{range .Functions}}
56+
{'type': 'function', 'function': {'name': '{{.Name}}', 'description': '{{.Description}}', 'parameters': {{toJson .Parameters}} }}
57+
{{end}}
58+
</tools>
59+
Use the following pydantic model json schema for each tool call you will make:
60+
{'title': 'FunctionCall', 'type': 'object', 'properties': {'arguments': {'title': 'Arguments', 'type': 'object'}, 'name': {'title': 'Name', 'type': 'string'}}, 'required': ['arguments', 'name']}
61+
For each function call return a json object with function name and arguments within <tool_call></tool_call> XML tags as follows:
62+
<tool_call>
63+
{'arguments': <args-dict>, 'name': <function-name>}
64+
</tool_call>
65+
<|im_end|>
66+
{{.Input -}}
67+
<|im_start|>assistant
68+
<tool_call>
69+
chat: |
70+
{{.Input -}}
71+
<|im_start|>assistant
72+
completion: |
73+
{{.Input}}
74+
context_size: 4096
75+
f16: true
76+
stopwords:
77+
- <|im_end|>
78+
- <dummy32000>
79+
- "\n</tool_call>"
80+
- "\n\n\n"
81+

0 commit comments

Comments
 (0)