Initial GPTQ model commit

README.md

@@ -53,21 +53,22 @@ Each separate quant is in a different branch.  See below for instructions on fet
 
 | Branch | Bits | Group Size | Act Order (desc_act) | File Size | ExLlama Compatible? | Made With | Description |
 | ------ | ---- | ---------- | -------------------- | --------- | ------------------- | --------- | ----------- |
-| main | 4 | 128 | False | 36.65 GB | True | AutoGPTQ | Most compatible option. Good inference speed in AutoGPTQ and GPTQ-for-LLaMa. Lower inference quality than other options. | 
+| main | 4 | None | True | 35.33 GB | True | AutoGPTQ | Most compatible option. Good inference speed in AutoGPTQ and GPTQ-for-LLaMa. Lower inference quality than other options. | 
+| gptq-3bit--1g-actorder_True | 3 | None | True | 26.78 GB | False | AutoGPTQ | 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_False | 3 | 128 | False | 28.03 GB | False | AutoGPTQ | 3-bit, with group size 128g but no act-order. Slightly higher VRAM requirements than 3-bit None. | 
+| gptq-3bit-128g-actorder_True | 3 | 128 | True | 28.03 GB | False | AutoGPTQ | 3-bit, with group size 128g and act-order. Higher quality than 128g-False but poor AutoGPTQ CUDA speed. | 
+| gptq-3bit-64g-actorder_True | 3 | 64 | True | 29.30 GB | False | AutoGPTQ | 3-bit, with group size 64g and act-order. Highest quality 3-bit option. Poor AutoGPTQ CUDA speed. | 
 | gptq-4bit-32g-actorder_True | 4 | 32 | True | 40.66 GB | True | AutoGPTQ | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. Poor AutoGPTQ CUDA speed. | 
 | gptq-4bit-64g-actorder_True | 4 | 64 | True | 37.99 GB | True | AutoGPTQ | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. | 
 | gptq-4bit-128g-actorder_True | 4 | 128 | True | 36.65 GB | True | AutoGPTQ | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. | 
-| gptq-8bit--1g-actorder_True | 8 | None | True | Processing, coming soon | False | AutoGPTQ | 8-bit, with Act Order. No group size, to lower VRAM requirements and to improve AutoGPTQ speed. | 
-| gptq-8bit-128g-actorder_False | 8 | 128 | False | Processing, coming soon | False | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and without Act Order to improve AutoGPTQ speed. | 
-| gptq-8bit-128g-actorder_True | 8 | 128 | True | Processing, coming soon | False | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. Poor AutoGPTQ CUDA speed. | 
-| gptq-8bit-64g-actorder_True | 8 | 64 | True | Processing, coming soon | False | AutoGPTQ | 8-bit, with group size 64g and Act Order for maximum inference quality. Poor AutoGPTQ CUDA speed. |
+| gptq-4bit-128g-actorder_False | 4 | 128 | False | 36.65 GB | True | AutoGPTQ | 4-bit, without Act Order and group size 128g. |
 
 ## How to download from branches
 
-- In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/airoboros-l2-70B-gpt4-1.4.1-GPTQ:gptq-4bit-32g-actorder_True`
+- In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/airoboros-l2-70B-gpt4-1.4.1-GPTQ:gptq-3bit--1g-actorder_True`
 - With Git, you can clone a branch with:
 ```
-git clone --branch gptq-4bit-32g-actorder_True https://huggingface.co/TheBloke/airoboros-l2-70B-gpt4-1.4.1-GPTQ`
+git clone --branch gptq-3bit--1g-actorder_True https://huggingface.co/TheBloke/airoboros-l2-70B-gpt4-1.4.1-GPTQ`
 ```
 - In Python Transformers code, the branch is the `revision` parameter; see below.
 
@@ -79,7 +80,7 @@ It is strongly recommended to use the text-generation-webui one-click-installers
 
 1. Click the **Model tab**.
 2. Under **Download custom model or LoRA**, enter `TheBloke/airoboros-l2-70B-gpt4-1.4.1-GPTQ`.
-  - To download from a specific branch, enter for example `TheBloke/airoboros-l2-70B-gpt4-1.4.1-GPTQ:gptq-4bit-32g-actorder_True`
+  - To download from a specific branch, enter for example `TheBloke/airoboros-l2-70B-gpt4-1.4.1-GPTQ:gptq-3bit--1g-actorder_True`
   - see Provided Files above for the list of branches for each option.
 3. Click **Download**.
 4. The model will start downloading. Once it's finished it will say "Done"
@@ -103,7 +104,7 @@ from transformers import AutoTokenizer, pipeline, logging
 from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
 
 model_name_or_path = "TheBloke/airoboros-l2-70B-gpt4-1.4.1-GPTQ"
-model_basename = "gptq_model-4bit-128g"
+model_basename = "gptq_model-4bit--1g"
 
 use_triton = False
 
@@ -121,7 +122,7 @@ model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
 To download from a specific branch, use the revision parameter, as in this example:
 
 model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
-        revision="gptq-4bit-32g-actorder_True",
+        revision="gptq-3bit--1g-actorder_True",
         model_basename=model_basename,
         use_safetensors=True,
         trust_remote_code=False,
@@ -198,10 +199,10 @@ Thank you to all my generous patrons and donaters!
 
 ### Overview
 
-Llama 2 version of https://huggingface.co/jondurbin/airoboros-70b-gpt4-1.4.1-qlora
-
-See that model card for all the details.
+Llama 2 70b fine tune using https://huggingface.co/datasets/jondurbin/airoboros-gpt4-1.4.1
 
+See the previous llama 65b model card for info:
+https://hf.co/jondurbin/airoboros-65b-gpt4-1.4
 
 ### Licence and usage restrictions