# mem0-embeddings-litellm-patch
This patch adds support for embedding model providers via [LiteLLM](https://github.com/BerriAI/litellm) to the [mem0](https://github.com/mem0-ai/mem0) framework.
## ✨ What It Does
- Integrates nearly all providers supported by LiteLLM as embedding backends in mem0
- Enables use of high-performance providers like **VoyageAI**, **Mistral**, **Groq**, and more
- Drop-in replacement for the existing embedding logic
## 🔧 Installation
You can install the patch via pip:
```bash
pip install mem0-embeddings-litellm-patch
````
This will patch the necessary `mem0.embeddings` modules automatically.
> **Note:** Make sure `mem0` and `litellm` are installed as dependencies. This package does not install them implicitly.
## 🧠 Requirements
* Python >= 3.8
* `mem0` >= 0.1.0
* `litellm` >= 1.0.0
## 💡 Usage
After installing this patch you can use all embedding providers available via litellm inncluding those currently not supported via mem0 natively.
## 📢 Why This Exists
The mem0 maintainers have not yet merged support for LiteLLM-based embeddings, despite it being a fast, extensible abstraction layer.
This patch bridges the gap until (or if) native support is added upstream. No need to fork and maintain a full project if you can just maintain the patch files instead am i right? :D No need to fork and maintain a full project if you can just maintain the patch files instead am i right? :D
## 📬 Feedback / Contributing
Feel free to fork or open issues. If the mem0 team integrates this feature officially, this package may be deprecated in favor of upstream support.
---
Licensed under the MIT License.
Raw data
{
"_id": null,
"home_page": null,
"name": "mem0-embeddings-litellm-patch",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": "Nocyphr <sgu@nocyphr.com>",
"keywords": "mem0, litellm, embeddings, ai, memory, patch",
"author": null,
"author_email": "Nocyphr <sgu@nocyphr.com>",
"download_url": "https://files.pythonhosted.org/packages/68/c5/8f4ec4c836da47e0f7e4e08cd79c99828b9c38eb55455ef2bcc42b74bf5e/mem0_embeddings_litellm_patch-1.0.9.tar.gz",
"platform": null,
"description": "# mem0-embeddings-litellm-patch\n\nThis patch adds support for embedding model providers via [LiteLLM](https://github.com/BerriAI/litellm) to the [mem0](https://github.com/mem0-ai/mem0) framework.\n\n## \u2728 What It Does\n\n- Integrates nearly all providers supported by LiteLLM as embedding backends in mem0\n- Enables use of high-performance providers like **VoyageAI**, **Mistral**, **Groq**, and more\n- Drop-in replacement for the existing embedding logic\n\n## \ud83d\udd27 Installation\n\nYou can install the patch via pip:\n\n```bash\npip install mem0-embeddings-litellm-patch\n````\n\nThis will patch the necessary `mem0.embeddings` modules automatically.\n\n> **Note:** Make sure `mem0` and `litellm` are installed as dependencies. This package does not install them implicitly.\n\n## \ud83e\udde0 Requirements\n\n* Python >= 3.8\n* `mem0` >= 0.1.0\n* `litellm` >= 1.0.0\n\n## \ud83d\udca1 Usage\n\nAfter installing this patch you can use all embedding providers available via litellm inncluding those currently not supported via mem0 natively. \n\n## \ud83d\udce2 Why This Exists\n\nThe mem0 maintainers have not yet merged support for LiteLLM-based embeddings, despite it being a fast, extensible abstraction layer.\nThis patch bridges the gap until (or if) native support is added upstream. No need to fork and maintain a full project if you can just maintain the patch files instead am i right? :D No need to fork and maintain a full project if you can just maintain the patch files instead am i right? :D\n\n## \ud83d\udcec Feedback / Contributing\n\nFeel free to fork or open issues. If the mem0 team integrates this feature officially, this package may be deprecated in favor of upstream support.\n\n---\n\nLicensed under the MIT License.\n\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Almost all known embedding model providers available via litellm patch",
"version": "1.0.9",
"project_urls": {
"Homepage": "https://github.com/nocyphr/mem0-embeddings-litellm-patch",
"Repository": "https://github.com/nocyphr/mem0-embeddings-litellm-patch"
},
"split_keywords": [
"mem0",
" litellm",
" embeddings",
" ai",
" memory",
" patch"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "bb7b223e80347cc728ba3fa307b46b377b8a8191afe32df3270ddf41388ee01d",
"md5": "3084fb8c32567def97a392d2ea5d6267",
"sha256": "a95e1f2c74e1020b1184ea429a1c6e35e290b7636dbbe4bdbb535533ffa9c7de"
},
"downloads": -1,
"filename": "mem0_embeddings_litellm_patch-1.0.9-py3-none-any.whl",
"has_sig": false,
"md5_digest": "3084fb8c32567def97a392d2ea5d6267",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 5362,
"upload_time": "2025-08-02T03:09:33",
"upload_time_iso_8601": "2025-08-02T03:09:33.459743Z",
"url": "https://files.pythonhosted.org/packages/bb/7b/223e80347cc728ba3fa307b46b377b8a8191afe32df3270ddf41388ee01d/mem0_embeddings_litellm_patch-1.0.9-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "68c58f4ec4c836da47e0f7e4e08cd79c99828b9c38eb55455ef2bcc42b74bf5e",
"md5": "c43598f2e71896b7b5e41430ce0b2e9d",
"sha256": "87dafb5d4deb3c353d2a843cfa4b62b64621a614b86146313ecff85a9b8b6048"
},
"downloads": -1,
"filename": "mem0_embeddings_litellm_patch-1.0.9.tar.gz",
"has_sig": false,
"md5_digest": "c43598f2e71896b7b5e41430ce0b2e9d",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 4544,
"upload_time": "2025-08-02T03:09:34",
"upload_time_iso_8601": "2025-08-02T03:09:34.692164Z",
"url": "https://files.pythonhosted.org/packages/68/c5/8f4ec4c836da47e0f7e4e08cd79c99828b9c38eb55455ef2bcc42b74bf5e/mem0_embeddings_litellm_patch-1.0.9.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-08-02 03:09:34",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "nocyphr",
"github_project": "mem0-embeddings-litellm-patch",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "mem0-embeddings-litellm-patch"
}
JSON
