Sweaterdog commited on
Commit
9e9a258
·
verified ·
1 Parent(s): 81add7b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +16 -16
README.md CHANGED
@@ -12,21 +12,21 @@ tags:
12
  - minecraft
13
  ---
14
 
15
- # 🧠 Andy-4-Preview 🧠
16
 
17
- Welcome to **Andy-4-Preview** – a revolutionary new model designed for playing Minecraft via the Mindcraft framework.
18
 
19
  This AI is crafted to push the boundaries of gameplay, reasoning, and multi-language summarization, making it one of the most versatile and powerful models available for Minecraft enthusiasts.
20
 
21
  # Overview
22
 
23
- Andy-4-Preview is an 8B parameter model built on the efficient Llama3.1 8B DeepSeek-R1 distill architecture.
24
 
25
  It has been meticulously trained over three weeks on a single RTX 3090 using two carefully curated datasets.
26
 
27
  The model underwent 2 epochs on the first dataset with a higher learning rate and 4 epochs on the second dataset with a much lower learning rate, ensuring a balanced and robust learning process.
28
 
29
- This training regimen, coupled with advanced techniques like manual learning rate adjustments and dynamic dataset modifications, has allowed Andy-4-Preview to outperform its competitors, including GPT-4o-mini, Claude 3.5 Haiku, Mineslayer v2, and the former leader, Andy-3.6.
30
 
31
  # How to install
32
 
@@ -45,17 +45,17 @@ If you do not have a dedicated GPU, there is a [guide in the Mindcraft discord](
45
 
46
  If you want a more custom install, go to the `files` tab on Huggingface, and download the quantization you want and the `Modelfile`
47
 
48
- Once downloaded, open `Modelfile` in a text editor, and change the `FROM` tag to the exact path of the GGUF weights you installed without the quotes, such as `C:\users\jimmy\downloads\Andy-4-preview.Q3_K_M.GGUF`
49
 
50
- Open the directory of your Modelfile in the terminal, or command prompt, and then run `ollama create Andy-4-preview -f Modelfile`, which will create `Andy-4-preview` as the model, based on the Modelfile, which includes some important information such as system prompt, chat template, context length, as well as sampling settings.
51
 
52
- For most people, using the direct HF method of downloading Andy-4-preview is recommended, but you will suffer from a shorter context window, and possibly sparatic text.
53
 
54
  For the full Andy-4 model, there will be a direct download on Ollama, so you won't have to do it this method.
55
 
56
  ## Installation notes
57
 
58
- Andy-4-preview, and Andy-4, will support a context window of up to `131072` tokens, but to run the Q5_K_M version with that context length, at stock settings on ollama, you would need an RTX 5090, and you would have 3GB of VRAM left over for Minecraft
59
 
60
  A context window of `8192` is smaller, but it can still allow for great conversations with the model, and since the model knows all of the commands by heart, you can cut the command docs from the system prompt to lower context usage.
61
 
@@ -96,7 +96,7 @@ ollama serve
96
  # Key Features
97
 
98
  - **Revolutionary Performance:** Specifically engineered for Minecraft gameplay, providing creative, strategic, and efficient in-game decision-making.
99
- - **Advanced Reasoning Capabilities:** If you include `Reason in your responses with <think> and </think>` in your prompt, or something similar, Andy-4-Preview will provide detailed reasoning to enhance performance, at the cost of speed.
100
  - **Multi-Language Summarization:** Capable of summarizing content in multiple languages, making it more efficient at remembering it's history.
101
  - **Building and Creativity:** Not only can it play Minecraft, but it also excels in constructing complex structures and solving intricate in-game challenges.
102
  - **Vast Knowledge Base:** Possesses extensive knowledge about Minecraft, including game mechanics, strategies, and creative builds.
@@ -105,11 +105,11 @@ ollama serve
105
 
106
  # Open Source and Licensing
107
 
108
- Andy-4-Preview is 100% open source and is licensed under the [Apache 2.0 License](LICENSE). We believe in transparency and community collaboration, and all source code and training details are available for review and contribution.
109
 
110
  ## LoRA Weights
111
 
112
- Access the LoRA weights for Andy-4-Preview here: [Andy-4-Preview LoRA Weights](https://huggingface.co/Sweaterdog/Andy-4-preview-LoRA)
113
 
114
  # Datasets
115
 
@@ -119,11 +119,11 @@ The model was trained on two distinct datasets:
119
 
120
  # Usage
121
 
122
- To integrate Andy-4-Preview with the Mindcraft framework, adjust your configuration settings as needed and follow the execution instructions provided separately.
123
 
124
  Customize the personality in the `conversing` profile section and configuration files to optimize performance for specific in-game tasks or environments.
125
 
126
- Andy-4-Preview is designed to seamlessly integrate with your existing Mindcraft setups, enhancing both gameplay and creative capabilities.
127
 
128
  # Mindcraft Project
129
 
@@ -132,7 +132,7 @@ Download and explore the Mindcraft project on GitHub:
132
 
133
  ## How to use
134
 
135
- On Mindcraft, ensure you put `ollama/` before the model name, this can be something like `ollama/hf.co/sweaterdog/andy-4-preview:q5_k_m`
136
 
137
  ## Notes on Quantization
138
 
@@ -143,7 +143,7 @@ For most cases, I suggest Q5_K_M or Q4_K_M, this is for 8GB GPUs, anything less
143
  # Disclaimer
144
 
145
  **Important Notice:**
146
- Andy-4-Preview is a preview model and, while it represents a significant advancement in AI-driven Minecraft gameplay, please be aware of the following:
147
  - **Performance Variability:** Due to its experimental nature, the model may not always deliver ultimate performance in every scenario.
148
  - **Ongoing Development:** This preview release is intended for early testing and community feedback. You might encounter occasional inconsistencies or limitations as further refinements are made.
149
 
@@ -154,7 +154,7 @@ Join our vibrant community for discussions, support, and feedback:
154
  - **Huggingface Page:** Explore more experimental models and projects: [Sweaterdog on Huggingface](https://huggingface.co/Sweaterdog)
155
  - **Discord Username:** `Sweaterdog`
156
 
157
- We welcome your feedback, suggestions, and contributions as we continue to improve Andy-4-Preview and push the boundaries of AI in gaming.
158
 
159
  # Acknowledgements
160
 
 
12
  - minecraft
13
  ---
14
 
15
+ # 🧠 Andy-4-base 🧠
16
 
17
+ Welcome to **Andy-4-base** – a revolutionary new model designed for playing Minecraft via the Mindcraft framework.
18
 
19
  This AI is crafted to push the boundaries of gameplay, reasoning, and multi-language summarization, making it one of the most versatile and powerful models available for Minecraft enthusiasts.
20
 
21
  # Overview
22
 
23
+ Andy-4-base is an 8B parameter model built on the efficient Llama3.1 8B DeepSeek-R1 distill architecture.
24
 
25
  It has been meticulously trained over three weeks on a single RTX 3090 using two carefully curated datasets.
26
 
27
  The model underwent 2 epochs on the first dataset with a higher learning rate and 4 epochs on the second dataset with a much lower learning rate, ensuring a balanced and robust learning process.
28
 
29
+ This training regimen, coupled with advanced techniques like manual learning rate adjustments and dynamic dataset modifications, has allowed Andy-4-base to outperform its competitors, including GPT-4o-mini, Claude 3.5 Haiku, Mineslayer v2, and the former leader, Andy-3.6.
30
 
31
  # How to install
32
 
 
45
 
46
  If you want a more custom install, go to the `files` tab on Huggingface, and download the quantization you want and the `Modelfile`
47
 
48
+ Once downloaded, open `Modelfile` in a text editor, and change the `FROM` tag to the exact path of the GGUF weights you installed without the quotes, such as `C:\users\jimmy\downloads\Andy-4-base.Q3_K_M.GGUF`
49
 
50
+ Open the directory of your Modelfile in the terminal, or command prompt, and then run `ollama create Andy-4-base -f Modelfile`, which will create `Andy-4-base` as the model, based on the Modelfile, which includes some important information such as system prompt, chat template, context length, as well as sampling settings.
51
 
52
+ For most people, using the direct HF method of downloading Andy-4-base is recommended, but you will suffer from a shorter context window, and possibly sparatic text.
53
 
54
  For the full Andy-4 model, there will be a direct download on Ollama, so you won't have to do it this method.
55
 
56
  ## Installation notes
57
 
58
+ Andy-4-base, and Andy-4, will support a context window of up to `131072` tokens, but to run the Q5_K_M version with that context length, at stock settings on ollama, you would need an RTX 5090, and you would have 3GB of VRAM left over for Minecraft
59
 
60
  A context window of `8192` is smaller, but it can still allow for great conversations with the model, and since the model knows all of the commands by heart, you can cut the command docs from the system prompt to lower context usage.
61
 
 
96
  # Key Features
97
 
98
  - **Revolutionary Performance:** Specifically engineered for Minecraft gameplay, providing creative, strategic, and efficient in-game decision-making.
99
+ - **Advanced Reasoning Capabilities:** If you include `Reason in your responses with <think> and </think>` in your prompt, or something similar, Andy-4-base will provide detailed reasoning to enhance performance, at the cost of speed.
100
  - **Multi-Language Summarization:** Capable of summarizing content in multiple languages, making it more efficient at remembering it's history.
101
  - **Building and Creativity:** Not only can it play Minecraft, but it also excels in constructing complex structures and solving intricate in-game challenges.
102
  - **Vast Knowledge Base:** Possesses extensive knowledge about Minecraft, including game mechanics, strategies, and creative builds.
 
105
 
106
  # Open Source and Licensing
107
 
108
+ Andy-4-base is 100% open source and is licensed under the [Apache 2.0 License](LICENSE). We believe in transparency and community collaboration, and all source code and training details are available for review and contribution.
109
 
110
  ## LoRA Weights
111
 
112
+ Access the LoRA weights for Andy-4-base here: [Andy-4-base LoRA Weights](https://huggingface.co/Sweaterdog/Andy-4-base-LoRA)
113
 
114
  # Datasets
115
 
 
119
 
120
  # Usage
121
 
122
+ To integrate Andy-4-base with the Mindcraft framework, adjust your configuration settings as needed and follow the execution instructions provided separately.
123
 
124
  Customize the personality in the `conversing` profile section and configuration files to optimize performance for specific in-game tasks or environments.
125
 
126
+ Andy-4-base is designed to seamlessly integrate with your existing Mindcraft setups, enhancing both gameplay and creative capabilities.
127
 
128
  # Mindcraft Project
129
 
 
132
 
133
  ## How to use
134
 
135
+ On Mindcraft, ensure you put `ollama/` before the model name, this can be something like `ollama/hf.co/sweaterdog/andy-4-base:q5_k_m`
136
 
137
  ## Notes on Quantization
138
 
 
143
  # Disclaimer
144
 
145
  **Important Notice:**
146
+ Andy-4-base is a preview model and, while it represents a significant advancement in AI-driven Minecraft gameplay, please be aware of the following:
147
  - **Performance Variability:** Due to its experimental nature, the model may not always deliver ultimate performance in every scenario.
148
  - **Ongoing Development:** This preview release is intended for early testing and community feedback. You might encounter occasional inconsistencies or limitations as further refinements are made.
149
 
 
154
  - **Huggingface Page:** Explore more experimental models and projects: [Sweaterdog on Huggingface](https://huggingface.co/Sweaterdog)
155
  - **Discord Username:** `Sweaterdog`
156
 
157
+ We welcome your feedback, suggestions, and contributions as we continue to improve Andy-4-base and push the boundaries of AI in gaming.
158
 
159
  # Acknowledgements
160