my-project-public

repository

loading code, commits, and activity

repositories

loading repo index

#1	---
#2	title: LM Studio
#3	---
#4
#5	To use LM Studio with Mem0, you'll need to have LM Studio running locally with its server enabled. LM Studio provides a way to run local LLMs with an OpenAI-compatible API.
#6
#7	## Usage
#8
#9	<CodeGroup>
#10	```python Python
#11	import os
#12	from mem0 import Memory
#13
#14	os.environ["OPENAI_API_KEY"] = "your-api-key" # used for embedding model
#15
#16	config = {
#17	"llm": {
#18	"provider": "lmstudio",
#19	"config": {
#20	"model": "lmstudio-community/Meta-Llama-3.1-70B-Instruct-GGUF/Meta-Llama-3.1-70B-Instruct-IQ2_M.gguf",
#21	"temperature": 0.2,
#22	"max_tokens": 2000,
#23	"lmstudio_base_url": "http://localhost:1234/v1", # default LM Studio API URL
#24	"lmstudio_response_format": {"type": "json_schema", "json_schema": {"type": "object", "schema": {}}},
#25	}
#26	}
#27	}
#28
#29	m = Memory.from_config(config)
#30	messages = [
#31	{"role": "user", "content": "I'm planning to watch a movie tonight. Any recommendations?"},
#32	{"role": "assistant", "content": "How about thriller movies? They can be quite engaging."},
#33	{"role": "user", "content": "I'm not a big fan of thriller movies but I love sci-fi movies."},
#34	{"role": "assistant", "content": "Got it! I'll avoid thriller recommendations and suggest sci-fi movies in the future."}
#35	]
#36	m.add(messages, user_id="alice", metadata={"category": "movies"})
#37	```
#38	</CodeGroup>
#39
#40	### Running Completely Locally
#41
#42	You can also use LM Studio for both LLM and embedding to run Mem0 entirely locally:
#43
#44	```python
#45	from mem0 import Memory
#46
#47	# No external API keys needed!
#48	config = {
#49	"llm": {
#50	"provider": "lmstudio"
#51	},
#52	"embedder": {
#53	"provider": "lmstudio"
#54	}
#55	}
#56
#57	m = Memory.from_config(config)
#58	messages = [
#59	{"role": "user", "content": "I'm planning to watch a movie tonight. Any recommendations?"},
#60	{"role": "assistant", "content": "How about thriller movies? They can be quite engaging."},
#61	{"role": "user", "content": "I'm not a big fan of thriller movies but I love sci-fi movies."},
#62	{"role": "assistant", "content": "Got it! I'll avoid thriller recommendations and suggest sci-fi movies in the future."}
#63	]
#64	m.add(messages, user_id="alice123", metadata={"category": "movies"})
#65	```
#66
#67	<Note>
#68	When using LM Studio for both LLM and embedding, make sure you have:
#69	1. An LLM model loaded for generating responses
#70	2. An embedding model loaded for vector embeddings
#71	3. The server enabled with the correct endpoints accessible
#72	</Note>
#73
#74	<Note>
#75	To use LM Studio, you need to:
#76	1. Download and install [LM Studio](https://lmstudio.ai/)
#77	2. Start a local server from the "Server" tab
#78	3. Set the appropriate `lmstudio_base_url` in your configuration (default is usually http://localhost:1234/v1)
#79	</Note>
#80
#81	## Config
#82
#83	All available parameters for the `lmstudio` config are present in [Master List of All Params in Config](../config).
#84

z6Mkq5mY3JWtxoxUobWcfNHm7AkRubgSWEZTkBVqZXJviFZ5/my-project-public