-
Notifications
You must be signed in to change notification settings - Fork 6
/
Copy pathhuggingface-space-catalog-default.yaml
179 lines (179 loc) · 7.7 KB
/
huggingface-space-catalog-default.yaml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
name: Hugging Face
entries:
- title: IDEFICS Playground
label: Idefics
short_description: This demo showcases IDEFICS, a open-access large visual language model.
long_description: >-
IDEFICS, short for Image-aware Decoder Enhanced à la Flamingo with Interleaved Cross-attentionS, is an 80-billion parameter open-access visual language model. It can process both image and text inputs, enabling capabilities like image-based question answering, visual content description, and storytelling with multiple images. IDEFICS is a reproduction of Deepmind's closed-source Flamingo model, built using publicly available data. The model variants in this demo have been fine-tuned for conversational settings.
tags:
- Gradio
- Space
- Idefics
git_url: "https://huggingface.co/spaces/HuggingFaceM4/idefics_playground"
is_prototype: true
is_huggingface_space: true
enabled: false
is_new: true
environment_variables:
cpu:
default: 2
description: "Number of CPUs"
memory:
default: 16
description: "Memory in GB"
gpu:
default: 0
description: "Number of GPUs"
HF_AUTH_TOKEN:
default: ""
description: "Hugging Face Token (HuggingFace->Settings->Access Tokens)"
required: true
tooltip: "Get your token from HuggingFace->Settings->Access Tokens"
- title: Code Llama Playground 13B
label: Llama-13B
short_description: Code Generator built on top of 13-billion parameter Code Llama model.
long_description: >-
Code Llama Playground showcases the 13-billion parameter Code Llama model, focused on text and code generation. It's primarily designed for code completion rather than instructional purposes or interactive chatting. Additional insights into the model's capabilities and design can be found in the associated blog post and paper.
image_path: >-
https://raw.githubusercontent.com/cloudera/HuggingFace-Spaces/master/images/codellama-playground.png
tags:
- Gradio
- Space
- Llama
- 13B
git_url: "https://huggingface.co/spaces/codellama/codellama-playground"
is_prototype: true
is_huggingface_space: true
is_new: true
environment_variables:
cpu:
default: 2
description: "Number of CPUs"
memory:
default: 16
description: "Memory in GB"
gpu:
default: 0
description: "Number of GPUs"
HF_TOKEN:
default: ""
description: "Hugging Face Token (HuggingFace->Settings->Access Tokens)"
required: true
tooltip: "Get your token from HuggingFace->Settings->Access Tokens"
- title: Mistral 7B Instruct
label: Mistral-7B
short_description: Chatbot using Mistral 7B model with features like Sliding Window Attention & Grouped Query Attention for faster inference.
long_description: >-
Mistral 7B Instruct showcases Mistral-7B-v0.1, Mistral AI's first Large Language Model (LLM), featuring a decoder-based architecture. Key features include Sliding Window Attention with an 8k context length, Grouped Query Attention (GQA) for faster inference, and a byte-fallback BPE tokenizer. The Mistral-7B-Instruct-v0.1 variant is instruction fine-tuned for chat-based interactions. Both models are available under the Apache 2.0 license. Further information is in the release blog post.
image_path: >-
https://raw.githubusercontent.com/cloudera/HuggingFace-Spaces/master/images/mistral-7B.png
tags:
- Gradio
- Space
- Mistral
- 7B
git_url: "https://huggingface.co/spaces/osanseviero/mistral-super-fast"
is_prototype: true
is_huggingface_space: true
is_new: true
environment_variables:
cpu:
default: 2
description: "Number of CPUs"
memory:
default: 16
description: "Memory in GB"
gpu:
default: 0
description: "Number of GPUs"
HF_TOKEN:
default: ""
description: "Hugging Face Token (HuggingFace->Settings->Access Tokens)"
required: true
tooltip: "Get your token from HuggingFace->Settings->Access Tokens"
HUGGING_FACE_HUB_TOKEN:
default: ""
description: "Hugging Face Token (HuggingFace->Settings->Access Tokens)"
required: true
tooltip: "Get your token from HuggingFace->Settings->Access Tokens"
- title: Chat with DeepSeek Coder 33B
label: DeepSeek-33B
short_description: Code Generator utilizing DeepSeek-Coder 33-billion parameter model that is fine-tuned for chat instructions.
long_description: >-
DeepSeek-33B-Chat features the DeepSeek-Coder model, a 33-billion parameter code model fine-tuned for chat instructions. This space is dedicated to demonstrating its capabilities in handling chat-based coding queries.
image_path: >-
https://raw.githubusercontent.com/cloudera/HuggingFace-Spaces/master/images/deepseek-33b-chat.png
tags:
- Gradio
- Space
- DeepSeek
- 33B
git_url: "https://huggingface.co/spaces/deepseek-ai/deepseek-coder-33b-instruct"
is_prototype: true
is_huggingface_space: true
is_new: true
environment_variables:
cpu:
default: 4
description: "Number of CPUs"
memory:
default: 32
description: "Memory in GB"
gpu:
default: 4
description: "Number of GPUs"
HF_TOKEN:
default: ""
description: "Hugging Face Token (HuggingFace->Settings->Access Tokens)"
required: true
tooltip: "Get your token from HuggingFace->Settings->Access Tokens"
HUGGING_FACE_HUB_TOKEN:
default: ""
description: "Hugging Face Token (HuggingFace->Settings->Access Tokens)"
required: true
tooltip: "Get your token from HuggingFace->Settings->Access Tokens"
- title: Can You Run It? LLM version
label: Run-llm-version
short_description: Pick the right GPU for your machine learning model.
long_description: >-
Can you run it? LLM version" is a tool designed to help users assess the feasibility of training or running large language models (LLMs). It provides guidance on the necessary GPU resources and quantization levels for different LLMs.
image_path: >-
https://raw.githubusercontent.com/cloudera/HuggingFace-Spaces/master/images/run-llm-version.png
tags:
- Streamlit
- Space
- LLM
git_url: "https://huggingface.co/spaces/Vokturz/can-it-run-llm"
is_prototype: true
is_huggingface_space: true
is_new: true
- title: LLMLingua-2
label: LLMLingua
short_description: Efficient and Faithful Task-Agnostic Prompt Compression via Data Distillation
long_description: >-
LLMLingua-2, a small-size yet powerful prompt compression method trained via data distillation from GPT-4 for token classification with a BERT-level encoder, excels in task-agnostic compression. It surpasses LLMLingua in handling out-of-domain data, offering 3x-6x faster performance
image_path: >-
https://raw.githubusercontent.com/cloudera/HuggingFace-Spaces/master/images/llmlingua.png
tags:
- Gradio
- Space
- LLM
git_url: "https://huggingface.co/spaces/microsoft/llmlingua-2"
is_prototype: true
is_huggingface_space: true
is_new: true
environment_variables:
cpu:
default: 2
description: "Number of CPUs"
memory:
default: 8
description: "Memory in GB"
gpu:
default: 0
description: "Number of GPUs"
HF_TOKEN:
default: ""
description: "Hugging Face Token with write permissions (HuggingFace->Settings->Access Tokens)"
required: true
tooltip: "Get your token with write permissions from HuggingFace->Settings->Access Tokens"