Update README.md

README.md

@@ -9,83 +9,70 @@ tags:
 - chatqa
 - llama-3
 - pytorch
+base_model: nvidia/Llama3-ChatQA-1.5-8B
 quantized_by: bartowski
+lm_studio:
+  param_count: 8b
+  use_case: RAG
+  release_date: 28-04-2024
+  model_creator: nvidia
+  prompt_template: chatQA
+  system_prompt: "This is a chat between a user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions based on the context. The assistant should also indicate when the answer cannot be found in the context."
+  base_model: llama
+  original_repo: nvidia/Llama3-ChatQA-1.5-8B
 ---
 
-## Llamacpp imatrix Quantizations of Llama-3-ChatQA-1.5-8B
+## 💫 Community Model> Llama 3 ChatQA 1.5 8B by NVIDIA
 
-Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b2777">b2777</a> for quantization.
+*👾 [LM Studio](https://lmstudio.ai) Community models highlights program. Highlighting new & noteworthy models by the community. Join the conversation on [Discord](https://discord.gg/aPQfnNkxGC)*.
 
-Original model: https://huggingface.co/nvidia/Llama3-ChatQA-1.5-8B
+**Model creator:** [nvidia](https://huggingface.co/nvidia)<br>
+**Original model**: [Llama3-ChatQA-1.5-8B](https://huggingface.co/nvidia/Llama3-ChatQA-1.5-8B)<br>
+**GGUF quantization:** provided by [bartowski](https://huggingface.co/bartowski) based on `llama.cpp` release [b2777](https://github.com/ggerganov/llama.cpp/releases/tag/b2777)<br>
 
-All quants made using imatrix option with dataset provided by Kalomaze [here](https://github.com/ggerganov/llama.cpp/discussions/5263#discussioncomment-8395384)
+## Model Summary:
+ChatQA 1.5 is a series of models trained to excel at RAG (retrieval augmented generation) tasks.<br>
+This model may work for general uses, but it primarily meant for use as a context sumarizer or context extraction.<br>
+Using the context provided after the system message, the model is able to provide contextual and accurate answers to queries.<br>
 
-## Prompt format
+## Prompt Template:
 
-```
-System: {System}
+For now, you'll need to make your own template. Choose the `LM Studio Blank Preset` in your LM Studio.
 
-{Context}
+Then, set the system prompt to whatever you'd like (check the recommended one below), and set the following values:<br>
+`System Message Prefix`: 'System: '<br>
+`User Message Prefix`: '\n\nUser: '<br>
+`User Message Suffix`: '\n\nAssistant: <|begin_of_text|>'<br>
 
-User: {Question}
+If you want to provide context, place that in the system message suffix like so:
 
-Assistant: {Response}
+`System Message Suffix`: '\n\n{context}'<br>
 
-User: {Question}
-
-Assistant:
+Under the hood, the model will see a prompt that's formatted like:
 
 ```
+System: This is a chat between a user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions based on the context. The assistant should also indicate when the answer cannot be found in the context.
 
-## Download a file (not the whole branch) from below:
-
-| Filename | Quant type | File Size | Description |
-| -------- | ---------- | --------- | ----------- |
-| [ChatQA-1.5-8B-Q8_0.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q8_0.gguf) | Q8_0 | 8.54GB | Extremely high quality, generally unneeded but max available quant. |
-| [ChatQA-1.5-8B-Q6_K.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q6_K.gguf) | Q6_K | 6.59GB | Very high quality, near perfect, *recommended*. |
-| [ChatQA-1.5-8B-Q5_K_M.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q5_K_M.gguf) | Q5_K_M | 5.73GB | High quality, *recommended*. |
-| [ChatQA-1.5-8B-Q5_K_S.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q5_K_S.gguf) | Q5_K_S | 5.59GB | High quality, *recommended*. |
-| [ChatQA-1.5-8B-Q4_K_M.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q4_K_M.gguf) | Q4_K_M | 4.92GB | Good quality, uses about 4.83 bits per weight, *recommended*. |
-| [ChatQA-1.5-8B-Q4_K_S.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q4_K_S.gguf) | Q4_K_S | 4.69GB | Slightly lower quality with more space savings, *recommended*. |
-| [ChatQA-1.5-8B-IQ4_NL.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ4_NL.gguf) | IQ4_NL | 4.67GB | Decent quality, slightly smaller than Q4_K_S with similar performance *recommended*. |
-| [ChatQA-1.5-8B-IQ4_XS.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ4_XS.gguf) | IQ4_XS | 4.44GB | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
-| [ChatQA-1.5-8B-Q3_K_L.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q3_K_L.gguf) | Q3_K_L | 4.32GB | Lower quality but usable, good for low RAM availability. |
-| [ChatQA-1.5-8B-Q3_K_M.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q3_K_M.gguf) | Q3_K_M | 4.01GB | Even lower quality. |
-| [ChatQA-1.5-8B-IQ3_M.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ3_M.gguf) | IQ3_M | 3.78GB | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
-| [ChatQA-1.5-8B-IQ3_S.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ3_S.gguf) | IQ3_S | 3.68GB | Lower quality, new method with decent performance, recommended over Q3_K_S quant, same size with better performance. |
-| [ChatQA-1.5-8B-Q3_K_S.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q3_K_S.gguf) | Q3_K_S | 3.66GB | Low quality, not recommended. |
-| [ChatQA-1.5-8B-IQ3_XS.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ3_XS.gguf) | IQ3_XS | 3.51GB | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
-| [ChatQA-1.5-8B-IQ3_XXS.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ3_XXS.gguf) | IQ3_XXS | 3.27GB | Lower quality, new method with decent performance, comparable to Q3 quants. |
-| [ChatQA-1.5-8B-Q2_K.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-Q2_K.gguf) | Q2_K | 3.17GB | Very low quality but surprisingly usable. |
-| [ChatQA-1.5-8B-IQ2_M.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ2_M.gguf) | IQ2_M | 2.94GB | Very low quality, uses SOTA techniques to also be surprisingly usable. |
-| [ChatQA-1.5-8B-IQ2_S.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ2_S.gguf) | IQ2_S | 2.75GB | Very low quality, uses SOTA techniques to be usable. |
-| [ChatQA-1.5-8B-IQ2_XS.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ2_XS.gguf) | IQ2_XS | 2.60GB | Very low quality, uses SOTA techniques to be usable. |
-| [ChatQA-1.5-8B-IQ2_XXS.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ2_XXS.gguf) | IQ2_XXS | 2.39GB | Lower quality, uses SOTA techniques to be usable. |
-| [ChatQA-1.5-8B-IQ1_M.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ1_M.gguf) | IQ1_M | 2.16GB | Extremely low quality, *not* recommended. |
-| [ChatQA-1.5-8B-IQ1_S.gguf](https://huggingface.co/bartowski/Llama-3-ChatQA-1.5-8B-GGUF/blob/main/ChatQA-1.5-8B-IQ1_S.gguf) | IQ1_S | 2.01GB | Extremely low quality, *not* recommended. |
+This is some context
 
-## Which file should I choose?
-
-A great write up with charts showing various performances is provided by Artefact2 [here](https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9)
-
-The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.
-
-If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.
+User: {Question}
 
-If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.
+Assistant: 
+```
 
-Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.
+nVidia also seems to recommend starting your query with "Please give a full and complete answer for the question."
 
-If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.
+## Technical Details
 
-If you want to get more into the weeds, you can check out this extremely useful feature chart:
+Llama3-ChatQA-1.5 excels at conversational question answering (QA) and retrieval-augmented generation (RAG). Llama3-ChatQA-1.5 is developed using an improved training recipe from ChatQA (1.0), and it is built on top of Llama-3 base model.<br>
+Specifically, more conversational QA data was used to enhance its tabular and arithmetic calculation capability.
 
-[llama.cpp feature matrix](https://github.com/ggerganov/llama.cpp/wiki/Feature-matrix)
+## Special thanks
 
-But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.
+🙏 Special thanks to [Georgi Gerganov](https://github.com/ggerganov) and the whole team working on [llama.cpp](https://github.com/ggerganov/llama.cpp/) for making all of this possible.
 
-These I-quants can also be used on CPU and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.
+🙏 Special thanks to [Kalomaze](https://github.com/kalomaze) for his dataset (linked [here](https://github.com/ggerganov/llama.cpp/discussions/5263)) that was used for calculating the imatrix for the IQ1_M and IQ2_XS quants, which makes them usable even at their tiny size!
 
-The I-quants are *not* compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.
+## Disclaimers
 
-Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski
+LM Studio is not the creator, originator, or owner of any Model featured in the Community Model Program. Each Community Model is created and provided by third parties. LM Studio does not endorse, support, represent or guarantee the completeness, truthfulness, accuracy, or reliability of any Community Model.  You understand that Community Models can produce content that might be offensive, harmful, inaccurate or otherwise inappropriate, or deceptive. Each Community Model is the sole responsibility of the person or entity who originated such Model. LM Studio may not monitor or control the Community Models and cannot, and does not, take responsibility for any such Model. LM Studio disclaims all warranties or guarantees about the accuracy, reliability or benefits of the Community Models.  LM Studio further disclaims any warranty that the Community Model will meet your requirements, be secure, uninterrupted or available at any time or location, or error-free, viruses-free, or that any errors will be corrected, or otherwise. You will be solely responsible for any damage resulting from your use of or access to the Community Models, your downloading of any Community Model, or use of any other Community Model provided by or through LM Studio.