[feat add]FastEmbed embedding for local embeddings

[lucifertrj]

Add: FastEmbed Embedding for the local embedding inference use cases

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Refactor (does not change functionality, e.g. code style improvements, linting)
• Documentation update

How Has This Been Tested?

As of now, I have tested FastEmbed Embeddings with mem0 Memory Config directly.

from mem0 import Memory
import os
os.environ['GOOGLE_API_KEY'] = "<api-key>"

config = {
    "llm": {
        "provider": "gemini",
        "config": {
            "model": "gemini-2.5-flash-lite",
            "temperature": 0.8,
        }
    },
    "vector_store": {
        "provider": "qdrant",
        "config": {
            "collection_name": "travel",
            "path": "/tmp/db",
            "embedding_model_dims": 768,
        }
    },
    "embedder": {
        "provider": "fastembed",
        "config": {
            "model": "jinaai/jina-embeddings-v2-base-en"
        }
    }
}
client = Memory.from_config(config)
messages = [
    {"role": "user", "content": "What is the must try food in Baroda"},
    {"role": "assistant", "content": "Sev Usal is must"},
    {"role": "user", "content": "I'm not into street food, I prefer Gujarati thalis."},
    {"role": "assistant", "content": "Head to Mandap in Baroda, it’s famous for authentic Gujarati thalis."},
]
result1 = client.add(messages, user_id="personal", metadata={"category": "food"})
print(result1)

I will push the testing scripts under the tests folder too.

Please delete options that are not relevant.

• Unit Test
• Test Script (please provide)

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules
• I have checked my code and corrected any misspellings

Maintainer Checklist

• closes #xxxx (Replace xxxx with the GitHub issue number)
• Made sure Checks passed

[CLAassistant]

All committers have signed the CLA.

[parshvadaftari]

Hey @lucifertrj Thank you for implementing this but can you please see for the failing tests?

[parshvadaftari]

Can you remove max_lenth from here? Fastembed doesn't support this param.

[lucifertrj]

can you check now

[parshvadaftari]

Hey @lucifertrj the tests are failing. Can you please check?

[parshvadaftari]

Can you resolve merge conflicts?

[lucifertrj]

Can you resolve merge conflicts?

I have resolved it. I am getting this Vercel Authorization required to deploy although I have already approved.

[parshvadaftari]

Vercel deployment is for the maintainers of the project, so that is fine and not an issue on your end.

[lucifertrj]

Vercel deployment is for the maintainers of the project, so that is fine and not an issue on your end.

Alright got it.

[lucifertrj]

this is test.py I used to test (PFA: Screenshot):

from mem0 import Memory
import os
os.environ['GOOGLE_API_KEY'] = ""

config = {
    "llm": {
        "provider": "gemini",
        "config": {
            "model": "gemini-2.5-flash-lite",
            "temperature": 0.8,
        }
    },
    "vector_store": {
        "provider": "qdrant",
        "config": {
            "collection_name": "devfestbaroda",
            "path": "/tmp/db",
            "embedding_model_dims": 768,
        }
    },
    "embedder": {
        "provider": "fastembed",
        "config": {
            "model": "jinaai/jina-embeddings-v2-base-en"
        }
    }
}
client = Memory.from_config(config)
messages = [
    {"role": "user", "content": "What is the must try food in Baroda"},
    {"role": "assistant", "content": "Sev Usal is must"},
    {"role": "user", "content": "I'm not into street food, I prefer Gujarati thalis."},
    {"role": "assistant", "content": "Head to Mandap in Baroda, it’s famous for authentic Gujarati thalis."},
]
result1 = client.add(messages, user_id="personal", metadata={"category": "food"})
print(result1)

[parshvadaftari]

@lucifertrj there's no support for the embedding_dims with the fastembed, how will the dimensions be decided and also what if I want to specify a specific dimension, that is something which should be either documented or implemented. Rest of the implementation looks good to me.

[lucifertrj]

@lucifertrj there's no support for the embedding_dims with the fastembed, how will the dimensions be decided and also what if I want to specify a specific dimension, that is something which should be documented. Rest of the implementation looks good to me.

In that case, they will have to use a specific vector store, mainly Qdrant. FastEmbed goes well in hand with Qdrant vector store. I can help out with documentation.

[parshvadaftari]

Okay cool it looks good to me!

[parshvadaftari]

Looks good to me!

[parshvadaftari]

Thanks a lot for your contribution @lucifertrj 🚀

[lucifertrj]

Thanks a lot for your contribution @lucifertrj 🚀

let's go. Let me add the documentation too...

[parshvadaftari]

Let me know once you raise the PR!🔥