Text Generation
Transformers
GGUF
English
llama
TheBloke commited on
Commit
445a2e6
1 Parent(s): b68e6c3

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +101 -21
README.md CHANGED
@@ -1,4 +1,5 @@
1
  ---
 
2
  datasets:
3
  - psmathur/orca_mini_v1_dataset
4
  - ehartford/dolphin
@@ -6,12 +7,24 @@ inference: false
6
  language:
7
  - en
8
  library_name: transformers
9
- license: llama2
10
  model_creator: Pankaj Mathur
11
- model_link: https://huggingface.co/psmathur/orca_mini_v3_70b
12
  model_name: Orca Mini v3 70B
13
  model_type: llama
14
  pipeline_tag: text-generation
 
 
 
 
 
 
 
 
 
 
 
 
 
15
  quantized_by: TheBloke
16
  ---
17
 
@@ -36,23 +49,25 @@ quantized_by: TheBloke
36
  - Model creator: [Pankaj Mathur](https://huggingface.co/psmathur)
37
  - Original model: [Orca Mini v3 70B](https://huggingface.co/psmathur/orca_mini_v3_70b)
38
 
 
39
  ## Description
40
 
41
  This repo contains GGUF format model files for [Pankaj Mathur's Orca Mini v3 70B](https://huggingface.co/psmathur/orca_mini_v3_70b).
42
 
 
43
  <!-- README_GGUF.md-about-gguf start -->
44
  ### About GGUF
45
 
46
- GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.
47
 
48
- The key benefit of GGUF is that it is a extensible, future-proof format which stores more information about the model as metadata. It also includes significantly improved tokenization code, including for the first time full support for special tokens. This should improve performance, especially with models that use new special tokens and implement custom prompt templates.
49
 
50
- Here are a list of clients and libraries that are known to support GGUF:
51
- * [llama.cpp](https://github.com/ggerganov/llama.cpp).
52
- * [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions.
53
- * [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with full GPU accel across multiple platforms and GPU architectures. Especially good for story telling.
54
- * [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI with GPU acceleration on both Windows (NVidia and AMD), and macOS.
55
  * [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
 
56
  * [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
57
  * [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
58
  * [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
@@ -61,9 +76,9 @@ Here are a list of clients and libraries that are known to support GGUF:
61
  <!-- repositories-available start -->
62
  ## Repositories available
63
 
 
64
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/orca_mini_v3_70B-GPTQ)
65
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/orca_mini_v3_70B-GGUF)
66
- * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference (deprecated)](https://huggingface.co/TheBloke/orca_mini_v3_70B-GGML)
67
  * [Pankaj Mathur's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/psmathur/orca_mini_v3_70b)
68
  <!-- repositories-available end -->
69
 
@@ -82,12 +97,21 @@ Here are a list of clients and libraries that are known to support GGUF:
82
  ```
83
 
84
  <!-- prompt-template end -->
 
 
 
 
 
 
 
 
 
85
  <!-- compatibility_gguf start -->
86
  ## Compatibility
87
 
88
- These quantised GGUF files are compatible with llama.cpp from August 21st 2023 onwards, as of commit [6381d4e110bd0ec02843a60bbeb8b6fc37a9ace9](https://github.com/ggerganov/llama.cpp/commit/6381d4e110bd0ec02843a60bbeb8b6fc37a9ace9)
89
 
90
- They are now also compatible with many third party UIs and libraries - please see the list at the top of the README.
91
 
92
  ## Explanation of quantisation methods
93
  <details>
@@ -160,21 +184,75 @@ del orca_mini_v3_70b.Q8_0.gguf-split-a orca_mini_v3_70b.Q8_0.gguf-split-b
160
  </details>
161
  <!-- README_GGUF.md-provided-files end -->
162
 
163
- <!-- README_GGUF.md-how-to-run start -->
164
- ## Example `llama.cpp` command
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
165
 
166
- Make sure you are using `llama.cpp` from commit [6381d4e110bd0ec02843a60bbeb8b6fc37a9ace9](https://github.com/ggerganov/llama.cpp/commit/6381d4e110bd0ec02843a60bbeb8b6fc37a9ace9) or later.
167
 
168
- For compatibility with older versions of llama.cpp, or for any third-party libraries or clients that haven't yet updated for GGUF, please use GGML files instead.
 
 
 
 
169
 
 
 
170
  ```
171
- ./main -t 10 -ngl 32 -m orca_mini_v3_70b.q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### System:\n{system_message}\n\n### User:\n{prompt}\n\n### Assistant:"
 
 
 
 
 
 
 
 
 
 
 
172
  ```
173
- Change `-t 10` to the number of physical CPU cores you have. For example if your system has 8 cores/16 threads, use `-t 8`. If offloading all layers to GPU, set `-t 1`.
174
 
175
  Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
176
 
177
- Change `-c 4096` to the desired sequence length for this model. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
178
 
179
  If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
180
 
@@ -231,10 +309,12 @@ For further support, and discussions on these models and AI in general, join us
231
 
232
  [TheBloke AI's Discord server](https://discord.gg/theblokeai)
233
 
234
- ## Thanks, and how to contribute.
235
 
236
  Thanks to the [chirper.ai](https://chirper.ai) team!
237
 
 
 
238
  I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
239
 
240
  If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
@@ -246,7 +326,7 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
246
 
247
  **Special thanks to**: Aemon Algiz.
248
 
249
- **Patreon special mentions**: Russ Johnson, J, alfie_i, Alex, NimbleBox.ai, Chadd, Mandus, Nikolai Manek, Ken Nordquist, ya boyyy, Illia Dulskyi, Viktor Bowallius, vamX, Iucharbius, zynix, Magnesian, Clay Pascal, Pierre Kircher, Enrico Ros, Tony Hughes, Elle, Andrey, knownsqashed, Deep Realms, Jerry Meng, Lone Striker, Derek Yates, Pyrater, Mesiah Bishop, James Bentley, Femi Adebogun, Brandon Frisco, SuperWojo, Alps Aficionado, Michael Dempsey, Vitor Caleffi, Will Dee, Edmond Seymore, usrbinkat, LangChain4j, Kacper Wikieł, Luke Pendergrass, John Detwiler, theTransient, Nathan LeClaire, Tiffany J. Kim, biorpg, Eugene Pentland, Stanislav Ovsiannikov, Fred von Graf, terasurfer, Kalila, Dan Guido, Nitin Borwankar, 阿明, Ai Maven, John Villwock, Gabriel Puliatti, Stephen Murray, Asp the Wyvern, danny, Chris Smitley, ReadyPlayerEmma, S_X, Daniel P. Andersen, Olakabola, Jeffrey Morgan, Imad Khwaja, Caitlyn Gatomon, webtim, Alicia Loh, Trenton Dambrowitz, Swaroop Kallakuri, Erik Bjäreholt, Leonard Tan, Spiking Neurons AB, Luke @flexchar, Ajan Kanaga, Thomas Belote, Deo Leter, RoA, Willem Michiel, transmissions 11, subjectnull, Matthew Berman, Joseph William Delisle, David Ziegler, Michael Davis, Johann-Peter Hartmann, Talal Aujan, senxiiz, Artur Olbinski, Rainer Wilmers, Spencer Kim, Fen Risland, Cap'n Zoog, Rishabh Srivastava, Michael Levine, Geoffrey Montalvo, Sean Connelly, Alexandros Triantafyllidis, Pieter, Gabriel Tamborski, Sam, Subspace Studios, Junyu Yang, Pedro Madruga, Vadim, Cory Kujawski, K, Raven Klaugh, Randy H, Mano Prime, Sebastain Graf, Space Cruiser
250
 
251
 
252
  Thank you to all my generous patrons and donaters!
 
1
  ---
2
+ base_model: https://huggingface.co/psmathur/orca_mini_v3_70b
3
  datasets:
4
  - psmathur/orca_mini_v1_dataset
5
  - ehartford/dolphin
 
7
  language:
8
  - en
9
  library_name: transformers
10
+ license: other
11
  model_creator: Pankaj Mathur
 
12
  model_name: Orca Mini v3 70B
13
  model_type: llama
14
  pipeline_tag: text-generation
15
+ prompt_template: '### System:
16
+
17
+ {system_message}
18
+
19
+
20
+ ### User:
21
+
22
+ {prompt}
23
+
24
+
25
+ ### Assistant:
26
+
27
+ '
28
  quantized_by: TheBloke
29
  ---
30
 
 
49
  - Model creator: [Pankaj Mathur](https://huggingface.co/psmathur)
50
  - Original model: [Orca Mini v3 70B](https://huggingface.co/psmathur/orca_mini_v3_70b)
51
 
52
+ <!-- description start -->
53
  ## Description
54
 
55
  This repo contains GGUF format model files for [Pankaj Mathur's Orca Mini v3 70B](https://huggingface.co/psmathur/orca_mini_v3_70b).
56
 
57
+ <!-- description end -->
58
  <!-- README_GGUF.md-about-gguf start -->
59
  ### About GGUF
60
 
61
+ GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp. GGUF offers numerous advantages over GGML, such as better tokenisation, and support for special tokens. It is also supports metadata, and is designed to be extensible.
62
 
63
+ Here is an incomplate list of clients and libraries that are known to support GGUF:
64
 
65
+ * [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
66
+ * [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
67
+ * [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
68
+ * [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
 
69
  * [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
70
+ * [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
71
  * [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
72
  * [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
73
  * [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
 
76
  <!-- repositories-available start -->
77
  ## Repositories available
78
 
79
+ * [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/orca_mini_v3_70B-AWQ)
80
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/orca_mini_v3_70B-GPTQ)
81
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/orca_mini_v3_70B-GGUF)
 
82
  * [Pankaj Mathur's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/psmathur/orca_mini_v3_70b)
83
  <!-- repositories-available end -->
84
 
 
97
  ```
98
 
99
  <!-- prompt-template end -->
100
+ <!-- licensing start -->
101
+ ## Licensing
102
+
103
+ The creator of the source model has listed its license as `other`, and this quantization has therefore used that same license.
104
+
105
+ As this model is based on Llama 2, it is also subject to the Meta Llama 2 license terms, and the license files for that are additionally included. It should therefore be considered as being claimed to be licensed under both licenses. I contacted Hugging Face for clarification on dual licensing but they do not yet have an official position. Should this change, or should Meta provide any feedback on this situation, I will update this section accordingly.
106
+
107
+ In the meantime, any questions regarding licensing, and in particular how these two licenses might interact, should be directed to the original model repository: [Pankaj Mathur's Orca Mini v3 70B](https://huggingface.co/psmathur/orca_mini_v3_70b).
108
+ <!-- licensing end -->
109
  <!-- compatibility_gguf start -->
110
  ## Compatibility
111
 
112
+ These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
113
 
114
+ They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
115
 
116
  ## Explanation of quantisation methods
117
  <details>
 
184
  </details>
185
  <!-- README_GGUF.md-provided-files end -->
186
 
187
+ <!-- README_GGUF.md-how-to-download start -->
188
+ ## How to download GGUF files
189
+
190
+ **Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
191
+
192
+ The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
193
+ - LM Studio
194
+ - LoLLMS Web UI
195
+ - Faraday.dev
196
+
197
+ ### In `text-generation-webui`
198
+
199
+ Under Download Model, you can enter the model repo: TheBloke/orca_mini_v3_70B-GGUF and below it, a specific filename to download, such as: orca_mini_v3_70b.q4_K_M.gguf.
200
+
201
+ Then click Download.
202
+
203
+ ### On the command line, including multiple files at once
204
+
205
+ I recommend using the `huggingface-hub` Python library:
206
+
207
+ ```shell
208
+ pip3 install huggingface-hub>=0.17.1
209
+ ```
210
+
211
+ Then you can download any individual model file to the current directory, at high speed, with a command like this:
212
+
213
+ ```shell
214
+ huggingface-cli download TheBloke/orca_mini_v3_70B-GGUF orca_mini_v3_70b.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
215
+ ```
216
+
217
+ <details>
218
+ <summary>More advanced huggingface-cli download usage</summary>
219
+
220
+ You can also download multiple files at once with a pattern:
221
+
222
+ ```shell
223
+ huggingface-cli download TheBloke/orca_mini_v3_70B-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
224
+ ```
225
+
226
+ For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
227
 
228
+ To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
229
 
230
+ ```shell
231
+ pip3 install hf_transfer
232
+ ```
233
+
234
+ And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
235
 
236
+ ```shell
237
+ HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/orca_mini_v3_70B-GGUF orca_mini_v3_70b.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
238
  ```
239
+
240
+ Windows CLI users: Use `set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1` before running the download command.
241
+ </details>
242
+ <!-- README_GGUF.md-how-to-download end -->
243
+
244
+ <!-- README_GGUF.md-how-to-run start -->
245
+ ## Example `llama.cpp` command
246
+
247
+ Make sure you are using `llama.cpp` from commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
248
+
249
+ ```shell
250
+ ./main -ngl 32 -m orca_mini_v3_70b.q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### System:\n{system_message}\n\n### User:\n{prompt}\n\n### Assistant:"
251
  ```
 
252
 
253
  Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
254
 
255
+ Change `-c 4096` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
256
 
257
  If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
258
 
 
309
 
310
  [TheBloke AI's Discord server](https://discord.gg/theblokeai)
311
 
312
+ ## Thanks, and how to contribute
313
 
314
  Thanks to the [chirper.ai](https://chirper.ai) team!
315
 
316
+ Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
317
+
318
  I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
319
 
320
  If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
 
326
 
327
  **Special thanks to**: Aemon Algiz.
328
 
329
+ **Patreon special mentions**: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
330
 
331
 
332
  Thank you to all my generous patrons and donaters!