Upload README.md

README.md

@@ -42,15 +42,12 @@ quantized_by: TheBloke
 <!-- description start -->
 # Description
 
-This repo contains **EXPERIMENTAL** GPTQ model files for [Mistral AI_'s Mixtral 8X7B v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1).
+This repo contains GPTQ model files for [Mistral AI_'s Mixtral 8X7B v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1).
 
-## Requires AutoGPTQ PR + transformers 4.36.0
-
-These files were made with, and will currently only work with, this AutoGPTQ PR: https://github.com/LaaZa/AutoGPTQ/tree/Mixtral-fix
-
-To test, please build AutoGPTQ from source using that PR.  You also need Transformers version 4.36.0, released December 11th.
-
-Transformers support has just arrived also via two PRs - and is expected in main Transformers + Optimum tomorrow (Dec 12th).
+Mixtral GPTQs currently require:
+* Transformers 4.36.0 or later
+* either, AutoGPTQ 0.6 compiled from source, or
+* Transformers 4.37.0.dev0 compiled from Github with: `pip3 install git+https://github.com/huggingface/transformers`
 
 Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
 
@@ -58,6 +55,7 @@ Multiple GPTQ parameter permutations are provided; see Provided Files below for
 <!-- repositories-available start -->
 ## Repositories available
 
+* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/mixtral-8x7b-v0.1-AWQ)
 * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ)
 * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GGUF)
 * [Mistral AI_'s original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1)
@@ -74,6 +72,16 @@ Multiple GPTQ parameter permutations are provided; see Provided Files below for
 <!-- prompt-template end -->
 
 
+
+<!-- README_GPTQ.md-compatible clients start -->
+## Known compatible clients / servers
+
+GPTQ models are currently supported on Linux (NVidia/AMD) and Windows (NVidia only). macOS users: please use GGUF models.
+
+Mixtral GPTQs currently have special requirements - see Description above.
+
+<!-- README_GPTQ.md-compatible clients end -->
+
 <!-- README_GPTQ.md-provided-files start -->
 ## Provided files, and GPTQ parameters
 
@@ -98,13 +106,13 @@ Most GPTQ files are made with AutoGPTQ. Mistral models are currently made with T
 
 | Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
 | ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
-| main | 4 | None | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 23.81 GB | No | 4-bit, with Act Order. No group size, to lower VRAM requirements. | 
-| gptq-4bit-128g-actorder_True | 4 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 24.70 GB | No | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. | 
-| gptq-4bit-32g-actorder_True | 4 | 32 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 27.42 GB | No | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. | 
-| gptq-3bit--1g-actorder_True | 3 | None | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 18.01 GB | No | 3-bit, with Act Order and no group size. Lowest possible VRAM requirements. May be lower quality than 3-bit 128g. | 
-| gptq-3bit-128g-actorder_True | 3 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 18.85 GB | No | 3-bit, with group size 128g and act-order. Higher quality than 128g-False. | 
-| gptq-8bit--1g-actorder_True | 8 | None | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 47.04 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. | 
-| gptq-8bit-128g-actorder_True | 8 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 48.10 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. |
+| [main](https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ/tree/main) | 4 | None | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 23.81 GB | No | 4-bit, with Act Order. No group size, to lower VRAM requirements. | 
+| [gptq-4bit-128g-actorder_True](https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ/tree/gptq-4bit-128g-actorder_True) | 4 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 24.70 GB | No | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. | 
+| [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 27.42 GB | No | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. | 
+| [gptq-3bit--1g-actorder_True](https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ/tree/gptq-3bit--1g-actorder_True) | 3 | None | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 18.01 GB | No | 3-bit, with Act Order and no group size. Lowest possible VRAM requirements. May be lower quality than 3-bit 128g. | 
+| [gptq-3bit-128g-actorder_True](https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ/tree/gptq-3bit-128g-actorder_True) | 3 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 18.85 GB | No | 3-bit, with group size 128g and act-order. Higher quality than 128g-False. | 
+| [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 47.04 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. | 
+| [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 48.10 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. |
 
 <!-- README_GPTQ.md-provided-files end -->
 
@@ -178,7 +186,11 @@ Note that using Git with HF repos is strongly discouraged. It will be much slowe
 <!-- README_GPTQ.md-text-generation-webui start -->
 ## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui)
 
-**NOTE**: This will only work with the AutoGPTQ loader, and only if you build AutoGPTQ from source using https://github.com/LaaZa/AutoGPTQ/tree/Mixtral
+**NOTE**: Requires:
+
+* Transformers 4.36.0, or Transformers 4.37.0.dev0 from Github
+* Either AutoGPTQ 0.6 compiled from source and `Loader: AutoGPTQ`,
+* or, `Loader: Transformers`, if you installed Transformers from Github: `pip3 install git+https://github.com/huggingface/transformers`
 
 Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
 
@@ -203,15 +215,21 @@ It is strongly recommended to use the text-generation-webui one-click-installers
 
 <!-- README_GPTQ.md-text-generation-webui end -->
 
+<!-- README_GPTQ.md-use-from-tgi start -->
+## Serving this model from Text Generation Inference (TGI)
+
+Not currently supported for Mixtral models.
+
+<!-- README_GPTQ.md-use-from-tgi end -->
 <!-- README_GPTQ.md-use-from-python start -->
 ## Python code example: inference from this GPTQ model
 
 ### Install the necessary packages
 
-Requires: Transformers 4.33.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.
+Requires: Transformers 4.37.0.dev0 from Github, Optimum 1.16.0 or later, and AutoGPTQ 0.5.1 or later.
 
 ```shell
-pip3 install --upgrade transformers optimum
+pip3 install --upgrade "git+https://github.com/huggingface/transformers" optimum
 # If using PyTorch 2.1 + CUDA 12.x:
 pip3 install --upgrade auto-gptq
 # or, if using PyTorch 2.1 + CUDA 11.x:
@@ -224,35 +242,28 @@ If you are using PyTorch 2.0, you will need to install AutoGPTQ from source. Lik
 pip3 uninstall -y auto-gptq
 git clone https://github.com/PanQiWei/AutoGPTQ
 cd AutoGPTQ
-git checkout v0.5.1
-pip3 install .
+DISABLE_QIGEN=1 pip3 install .
 ```
 
 ### Example Python code
 
 ```python
-from transformers import AutoTokenizer
-from auto_gptq import AutoGPTQForCausalLM
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
 
 model_name_or_path = "TheBloke/Mixtral-8x7B-v0.1-GPTQ"
-
-model_name_or_path = args.model_dir
 # To use a different branch, change revision
-# For example: revision="gptq-4bit-32g-actorder_True"
-model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
-        model_basename="model",
-        use_safetensors=True,
-        trust_remote_code=False,
-        device="cuda:0",
-        use_triton=False,
-        disable_exllama=True,
-        disable_exllamav2=True,
-        quantize_config=None)
-
-tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True, trust_remote_code=False)
-
-prompt = "Tell me about AI"
-prompt_template=f'''{prompt}'''
+# For example: revision="gptq-4bit-128g-actorder_True"
+model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
+                                             device_map="auto",
+                                             trust_remote_code=False,
+                                             revision="main")
+
+tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
+
+prompt = "Write a story about llamas"
+system_message = "You are a story writing assistant"
+prompt_template=f'''{prompt}
+'''
 
 print("\n\n*** Generate:")
 
@@ -260,9 +271,31 @@ input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
 output = model.generate(inputs=input_ids, temperature=0.7, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=512)
 print(tokenizer.decode(output[0]))
 
+# Inference can also be done using transformers' pipeline
+
+print("*** Pipeline:")
+pipe = pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    max_new_tokens=512,
+    do_sample=True,
+    temperature=0.7,
+    top_p=0.95,
+    top_k=40,
+    repetition_penalty=1.1
+)
+
+print(pipe(prompt_template)[0]['generated_text'])
 ```
 <!-- README_GPTQ.md-use-from-python end -->
 
+<!-- README_GPTQ.md-compatibility start -->
+## Compatibility
+
+The files provided are tested to work with AutoGPTQ 0.6 (compiled from source) and Transformers 4.37.0 (installed from Github).
+
+<!-- README_GPTQ.md-compatibility end -->
 
 <!-- footer start -->
 <!-- 200823 -->