metadata

base_model: mixedbread-ai/mxbai-embed-large-v1
inference: false
language:
  - en
license: apache-2.0
model_creator: mixedbread-ai
model_name: mxbai-embed-large-v1
model_type: bert
quantized_by: ChristianAzinn
library_name: sentence-transformers
pipeline_tag: feature-extraction
tags:
  - mteb
  - transformers
  - transformers.js
  - gguf

mxbai-embed-large-v1-gguf

Model creator: MixedBread AI

Original model: mxbai-embed-large-v1

Original Description

This is our base sentence embedding model. It was trained using AnglE loss on our high-quality large scale data. It achieves SOTA performance on BERT-large scale. Find out more in our blog post.

Description

This repo contains GGUF format files for the mxbai-embed-large-v1 embedding model.

These files were converted and quantized with llama.cpp PR 5500, commit 34aa045de, on a consumer RTX 4090.

This model supports up to 512 tokens of context.

Compatibility

These files are compatible with llama.cpp as of commit 4524290e8, as well as LM Studio as of version 0.2.19.

Meta-information

Explanation of quantisation methods

Click to see details

The methods available are: * 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) * 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. * 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. * GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw * 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 Refer to the Provided Files table below to see what files use which methods, and how.

Provided Files

Name	Quant method	Bits	Size	Use case
mxbai-embed-large-v1.Q2_K.gguf	Q2_K	2	144 MB	smallest, significant quality loss - not recommended for most purposes
mxbai-embed-large-v1.Q3_K_S.gguf	Q3_K_S	3	160 MB	very small, high quality loss
mxbai-embed-large-v1.Q3_K_M.gguf	Q3_K_M	3	181 MB	very small, high quality loss
mxbai-embed-large-v1.Q3_K_L.gguf	Q3_K_L	3	198 MB	small, substantial quality loss
mxbai-embed-large-v1.Q4_0.gguf	Q4_0	4	200 MB	legacy; small, very high quality loss - prefer using Q3_K_M
mxbai-embed-large-v1.Q4_K_S.gguf	Q4_K_S	4	203 MB	small, greater quality loss
mxbai-embed-large-v1.Q4_K_M.gguf	Q4_K_M	4	216 MB	medium, balanced quality - recommended
mxbai-embed-large-v1.Q5_0.gguf	Q5_0	5	237 MB	legacy; medium, balanced quality - prefer using Q4_K_M
mxbai-embed-large-v1.Q5_K_S.gguf	Q5_K_S	5	237 MB	large, low quality loss - recommended
mxbai-embed-large-v1.Q5_K_M.gguf	Q5_K_M	5	246 MB	large, very low quality loss - recommended
mxbai-embed-large-v1.Q6_K.gguf	Q6_K	6	278 MB	very large, extremely low quality loss
mxbai-embed-large-v1.Q8_0.gguf	Q8_0	8	358 MB	very large, extremely low quality loss - recommended
mxbai-embed-large-v1.Q8_0.gguf	FP16	16	670 MB	enormous, pretty much the original model - not recommended
mxbai-embed-large-v1.Q8_0.gguf	FP32	32	1.34 GB	enormous, pretty much the original model - not recommended

Examples

Example Usage with `llama.cpp`

To compute a single embedding, build llama.cpp and run:

./embedding -ngl 99 -m [filepath-to-gguf].gguf -p 'search_query: What is TSNE?'

You can also submit a batch of texts to embed, as long as the total number of tokens does not exceed the context length. Only the first three embeddings are shown by the embedding example.

texts.txt:

search_query: What is TSNE?
search_query: Who is Laurens Van der Maaten?

Compute multiple embeddings:

./embedding -ngl 99 -m [filepath-to-gguf].gguf -f texts.txt

Example Usage with LM Studio

Download the 0.2.19 beta build from here: Windows MacOS Linux

Once installed, open the app. The home should look like this:

Search for either "ChristianAzinn" in the main search bar or go to the "Search" tab on the left menu and search the name there.

Select your model from those that appear (this example uses bge-small-en-v1.5-gguf) and select which quantization you want to download. Since this model is pretty small, I recommend Q8_0, if not f16/32. Generally, the lower you go in the list (or the bigger the number gets), the larger the file and the better the performance.

You will see a green checkmark and the word "Downloaded" once the model has successfully downloaded, which can take some time depending on your network speeds.

Once this model is finished downloading, navigate to the "Local Server" tab on the left menu and open the loader for text embedding models. This loader does not appear before version 0.2.19, so ensure you downloaded the correct version.

Select the model you just downloaded from the dropdown that appears to load it. You may need to play with configuratios in the right-side menu, such as GPU offload if it doesn't fit entirely into VRAM.

All that's left to do is to hit the "Start Server" button:

And if you see text like that shown below in the console, you're good to go! You can use this as a drop-in replacement for the OpenAI embeddings API in any application that requires it, or you can query the endpoint directly to test it out.

Example curl request to the API endpoint:

curl http://localhost:1234/v1/embeddings \
  -H "Content-Type: application/json" \
  -d '{
    "input": "Your text string goes here",
    "model": "model-identifier-here"
  }'

For more information, see the LM Studio text embedding documentation.

Acknowledgements

Thanks to the LM Studio team and everyone else working on open-source AI.

This README is inspired by that of nomic-ai-embed-text-v1.5-GGUF, another excellent embedding model, and those of the legendary TheBloke.