TheBloke commited on
Commit
e10d393
1 Parent(s): 375c4c9

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +300 -0
README.md ADDED
@@ -0,0 +1,300 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: mistralai/Mistral-7B-v0.1
3
+ inference: false
4
+ license: apache-2.0
5
+ model_creator: Mistral AI
6
+ model_name: Mistral 7B v0.1
7
+ model_type: mistral
8
+ pipeline_tag: text-generation
9
+ prompt_template: '{prompt}
10
+
11
+ '
12
+ quantized_by: TheBloke
13
+ tags:
14
+ - pretrained
15
+ ---
16
+
17
+ <!-- header start -->
18
+ <!-- 200823 -->
19
+ <div style="width: auto; margin-left: auto; margin-right: auto">
20
+ <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
21
+ </div>
22
+ <div style="display: flex; justify-content: space-between; width: 100%;">
23
+ <div style="display: flex; flex-direction: column; align-items: flex-start;">
24
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
25
+ </div>
26
+ <div style="display: flex; flex-direction: column; align-items: flex-end;">
27
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
28
+ </div>
29
+ </div>
30
+ <div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
31
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
32
+ <!-- header end -->
33
+
34
+ # Mistral 7B v0.1 - GGUF
35
+ - Model creator: [Mistral AI](https://huggingface.co/mistralai)
36
+ - Original model: [Mistral 7B v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1)
37
+
38
+ <!-- description start -->
39
+ ## Description
40
+
41
+ This repo contains GGUF format model files for [Mistral AI's Mistral 7B v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1).
42
+
43
+ <!-- description end -->
44
+ <!-- README_GGUF.md-about-gguf start -->
45
+ ### About GGUF
46
+
47
+ 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.
48
+
49
+ Here is an incomplate list of clients and libraries that are known to support GGUF:
50
+
51
+ * [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
52
+ * [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.
53
+ * [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.
54
+ * [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
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
+ * [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.
57
+ * [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
58
+ * [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
59
+ * [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
60
+
61
+ <!-- README_GGUF.md-about-gguf end -->
62
+ <!-- repositories-available start -->
63
+ ## Repositories available
64
+
65
+ * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF)
66
+ * [Mistral AI's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/mistralai/Mistral-7B-v0.1)
67
+ <!-- repositories-available end -->
68
+
69
+ <!-- prompt-template start -->
70
+ ## Prompt template: None
71
+
72
+ ```
73
+ {prompt}
74
+
75
+ ```
76
+
77
+ <!-- prompt-template end -->
78
+
79
+
80
+ <!-- compatibility_gguf start -->
81
+ ## Compatibility
82
+
83
+ These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
84
+
85
+ They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
86
+
87
+ ## Explanation of quantisation methods
88
+ <details>
89
+ <summary>Click to see details</summary>
90
+
91
+ The new methods available are:
92
+ * GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
93
+ * GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
94
+ * GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
95
+ * GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
96
+ * GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
97
+
98
+ Refer to the Provided Files table below to see what files use which methods, and how.
99
+ </details>
100
+ <!-- compatibility_gguf end -->
101
+
102
+ <!-- README_GGUF.md-provided-files start -->
103
+ ## Provided files
104
+
105
+ | Name | Quant method | Bits | Size | Max RAM required | Use case |
106
+ | ---- | ---- | ---- | ---- | ---- | ----- |
107
+ | [mistral-7b-v0.1.Q2_K.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q2_K.gguf) | Q2_K | 2 | 3.08 GB| 5.58 GB | smallest, significant quality loss - not recommended for most purposes |
108
+ | [mistral-7b-v0.1.Q3_K_S.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q3_K_S.gguf) | Q3_K_S | 3 | 3.16 GB| 5.66 GB | very small, high quality loss |
109
+ | [mistral-7b-v0.1.Q3_K_M.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q3_K_M.gguf) | Q3_K_M | 3 | 3.52 GB| 6.02 GB | very small, high quality loss |
110
+ | [mistral-7b-v0.1.Q3_K_L.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q3_K_L.gguf) | Q3_K_L | 3 | 3.82 GB| 6.32 GB | small, substantial quality loss |
111
+ | [mistral-7b-v0.1.Q4_0.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q4_0.gguf) | Q4_0 | 4 | 4.11 GB| 6.61 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
112
+ | [mistral-7b-v0.1.Q4_K_S.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q4_K_S.gguf) | Q4_K_S | 4 | 4.14 GB| 6.64 GB | small, greater quality loss |
113
+ | [mistral-7b-v0.1.Q4_K_M.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q4_K_M.gguf) | Q4_K_M | 4 | 4.37 GB| 6.87 GB | medium, balanced quality - recommended |
114
+ | [mistral-7b-v0.1.Q5_0.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q5_0.gguf) | Q5_0 | 5 | 5.00 GB| 7.50 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
115
+ | [mistral-7b-v0.1.Q5_K_S.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q5_K_S.gguf) | Q5_K_S | 5 | 5.00 GB| 7.50 GB | large, low quality loss - recommended |
116
+ | [mistral-7b-v0.1.Q5_K_M.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q5_K_M.gguf) | Q5_K_M | 5 | 5.13 GB| 7.63 GB | large, very low quality loss - recommended |
117
+ | [mistral-7b-v0.1.Q6_K.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q6_K.gguf) | Q6_K | 6 | 5.94 GB| 8.44 GB | very large, extremely low quality loss |
118
+ | [mistral-7b-v0.1.Q8_0.gguf](https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF/blob/main/mistral-7b-v0.1.Q8_0.gguf) | Q8_0 | 8 | 7.70 GB| 10.20 GB | very large, extremely low quality loss - not recommended |
119
+
120
+ **Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
121
+
122
+
123
+
124
+ <!-- README_GGUF.md-provided-files end -->
125
+
126
+ <!-- README_GGUF.md-how-to-download start -->
127
+ ## How to download GGUF files
128
+
129
+ **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.
130
+
131
+ The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
132
+ - LM Studio
133
+ - LoLLMS Web UI
134
+ - Faraday.dev
135
+
136
+ ### In `text-generation-webui`
137
+
138
+ Under Download Model, you can enter the model repo: TheBloke/Mistral-7B-v0.1-GGUF and below it, a specific filename to download, such as: mistral-7b-v0.1.Q4_K_M.gguf.
139
+
140
+ Then click Download.
141
+
142
+ ### On the command line, including multiple files at once
143
+
144
+ I recommend using the `huggingface-hub` Python library:
145
+
146
+ ```shell
147
+ pip3 install huggingface-hub
148
+ ```
149
+
150
+ Then you can download any individual model file to the current directory, at high speed, with a command like this:
151
+
152
+ ```shell
153
+ huggingface-cli download TheBloke/Mistral-7B-v0.1-GGUF mistral-7b-v0.1.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
154
+ ```
155
+
156
+ <details>
157
+ <summary>More advanced huggingface-cli download usage</summary>
158
+
159
+ You can also download multiple files at once with a pattern:
160
+
161
+ ```shell
162
+ huggingface-cli download TheBloke/Mistral-7B-v0.1-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
163
+ ```
164
+
165
+ 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).
166
+
167
+ To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
168
+
169
+ ```shell
170
+ pip3 install hf_transfer
171
+ ```
172
+
173
+ And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
174
+
175
+ ```shell
176
+ HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Mistral-7B-v0.1-GGUF mistral-7b-v0.1.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
177
+ ```
178
+
179
+ Windows Command Line users: You can set the environment variable by running `set HF_HUB_ENABLE_HF_TRANSFER=1` before the download command.
180
+ </details>
181
+ <!-- README_GGUF.md-how-to-download end -->
182
+
183
+ <!-- README_GGUF.md-how-to-run start -->
184
+ ## Example `llama.cpp` command
185
+
186
+ Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
187
+
188
+ ```shell
189
+ ./main -ngl 32 -m mistral-7b-v0.1.Q4_K_M.gguf --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "{prompt}"
190
+ ```
191
+
192
+ Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
193
+
194
+ Change `-c 2048` 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.
195
+
196
+ If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
197
+
198
+ For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
199
+
200
+ ## How to run in `text-generation-webui`
201
+
202
+ Further instructions here: [text-generation-webui/docs/llama.cpp.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp.md).
203
+
204
+ ## How to run from Python code
205
+
206
+ You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries.
207
+
208
+ ### How to load this model in Python code, using ctransformers
209
+
210
+ #### First install the package
211
+
212
+ Run one of the following commands, according to your system:
213
+
214
+ ```shell
215
+ # Base ctransformers with no GPU acceleration
216
+ pip install ctransformers
217
+ # Or with CUDA GPU acceleration
218
+ pip install ctransformers[cuda]
219
+ # Or with AMD ROCm GPU acceleration (Linux only)
220
+ CT_HIPBLAS=1 pip install ctransformers --no-binary ctransformers
221
+ # Or with Metal GPU acceleration for macOS systems only
222
+ CT_METAL=1 pip install ctransformers --no-binary ctransformers
223
+ ```
224
+
225
+ #### Simple ctransformers example code
226
+
227
+ ```python
228
+ from ctransformers import AutoModelForCausalLM
229
+
230
+ # Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
231
+ llm = AutoModelForCausalLM.from_pretrained("TheBloke/Mistral-7B-v0.1-GGUF", model_file="mistral-7b-v0.1.Q4_K_M.gguf", model_type="mistral", gpu_layers=50)
232
+
233
+ print(llm("AI is going to"))
234
+ ```
235
+
236
+ ## How to use with LangChain
237
+
238
+ Here are guides on using llama-cpp-python and ctransformers with LangChain:
239
+
240
+ * [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
241
+ * [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
242
+
243
+ <!-- README_GGUF.md-how-to-run end -->
244
+
245
+ <!-- footer start -->
246
+ <!-- 200823 -->
247
+ ## Discord
248
+
249
+ For further support, and discussions on these models and AI in general, join us at:
250
+
251
+ [TheBloke AI's Discord server](https://discord.gg/theblokeai)
252
+
253
+ ## Thanks, and how to contribute
254
+
255
+ Thanks to the [chirper.ai](https://chirper.ai) team!
256
+
257
+ Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
258
+
259
+ 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.
260
+
261
+ 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.
262
+
263
+ Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
264
+
265
+ * Patreon: https://patreon.com/TheBlokeAI
266
+ * Ko-Fi: https://ko-fi.com/TheBlokeAI
267
+
268
+ **Special thanks to**: Aemon Algiz.
269
+
270
+ **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
271
+
272
+
273
+ Thank you to all my generous patrons and donaters!
274
+
275
+ And thank you again to a16z for their generous grant.
276
+
277
+ <!-- footer end -->
278
+
279
+ <!-- original-model-card start -->
280
+ # Original model card: Mistral AI's Mistral 7B v0.1
281
+
282
+
283
+ # Model Card for Mistral-7B-v0.1
284
+
285
+ The Mistral-7B-v0.1 Large Language Model (LLM) is a pretrained generative text model with 7 billion parameters.
286
+ Mistral-7B-v0.1 outperforms Llama 2 13B on all benchmarks we tested.
287
+
288
+ For full details of this model please read our [Release blog post](https://mistral.ai/news/announcing-mistral-7b/)
289
+
290
+ ## Model Architecture
291
+ Mistral-7B-v0.1 is a transformer model, with the following architecture choices:
292
+ - Grouped-Query Attention
293
+ - Sliding-Window Attention
294
+ - Byte-fallback BPE tokenizer
295
+
296
+ ## The Mistral AI Team
297
+
298
+ Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lélio Renard Lavaud, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.
299
+
300
+ <!-- original-model-card end -->