JDSS/mem0
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(oss): port v3 pipeline with hybrid search, entity extraction, and additive scoring (#4805)

Browse Source 
Co-authored-by: Soumil Rathi <soumilrathi@gmail.com>
Co-authored-by: Saket Aryan <saketaryan2002@gmail.com>
Co-authored-by: chaithanyak42 <chaithanya.kumar42a@gmail.com>
Co-authored-by: kartik-mem0 <kartik.labhshetwar@mem0.ai>
This commit is contained in:
soumil-rathi
2026-04-14 05:30:58 -07:00
committed by GitHub
parent 57f944e18a
commit a488e19044
120 changed files with 10107 additions and 17135 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+10 -1
View File
@@ -88,6 +88,13 @@ Install the sdk via pip:
pip install mem0ai
```
For enhanced hybrid search with BM25 keyword matching and entity extraction, install with NLP support:
```bash
pip install mem0ai[nlp]
python -m spacy download en_core_web_sm
```
Install sdk via npm:
```bash
npm install mem0ai
@@ -109,7 +116,9 @@ See the [CLI documentation](https://docs.mem0.ai/platform/cli) for the full comm
### Basic Usage
Mem0 requires an LLM to function, with `gpt-4.1-nano-2025-04-14 from OpenAI as the default. However, it supports a variety of LLMs; for details, refer to our [Supported LLMs documentation](https://docs.mem0.ai/components/llms/overview).
Mem0 requires an LLM to function, with `gpt-4.1-nano-2025-04-14` from OpenAI as the default. However, it supports a variety of LLMs; for details, refer to our [Supported LLMs documentation](https://docs.mem0.ai/components/llms/overview).
Mem0 uses `text-embedding-3-small` from OpenAI as the default embedding model. For best results with hybrid search (semantic + keyword + entity boosting), we recommend using at least [Qwen 600M](https://huggingface.co/Alibaba-NLP/gte-Qwen2-1.5B-instruct) or a comparable embedding model. See [Supported Embeddings](https://docs.mem0.ai/components/embedders/overview) for configuration details.
First step is to instantiate the memory: