This is legacy documentation for Mem0 v0.x. For the latest FAQs, please refer to v1.0 Beta FAQs.

​
General Questions

​
What is Mem0 v0.x?

Mem0 v0.x is the legacy version of Mem0’s memory layer for LLMs. While still functional, it lacks the advanced features and optimizations available in v1.0 Beta.

​
Should I upgrade to v1.0 Beta?

Yes! v1.0 Beta offers significant improvements:
  • Enhanced filtering with logical operators
  • Reranking support for better search relevance
  • Improved async performance
  • Standardized API responses
  • Better error handling
See our migration guide for upgrade instructions.

​
Is v0.x still supported?

v0.x receives minimal maintenance but no new features. We recommend upgrading to v1.0 Beta for the latest improvements and active support.

​
API Questions

​
Why do I get different response formats?

In v0.x, response format depends on the output_format parameter: 
# v1.0 format (list)
result = m.add("memory", user_id="alice", output_format="v1.0")
# Returns: [{"id": "...", "memory": "...", "event": "ADD"}]

# v1.1 format (dict)
result = m.add("memory", user_id="alice", output_format="v1.1")
# Returns: {"results": [{"id": "...", "memory": "...", "event": "ADD"}]}
Solution: Always use output_format="v1.1" for consistency.

​
How do I handle both response formats?

def normalize_response(result):
    """Normalize v0.x response formats"""
    if isinstance(result, list):
        return {"results": result}
    return result

# Usage
result = m.add("memory", user_id="alice")
normalized = normalize_response(result)
for memory in normalized["results"]:
    print(memory["memory"])

​
Can I use async in v0.x?

Yes, but it’s optional and less optimized: 
# Optional async mode
result = m.add("memory", user_id="alice", async_mode=True)

# Or use AsyncMemory
from mem0 import AsyncMemory
async_m = AsyncMemory()
result = await async_m.add("memory", user_id="alice")

​
Configuration Questions

​
What vector stores work with v0.x?

v0.x supports most vector stores:
  • Qdrant
  • Chroma
  • Pinecone
  • Weaviate
  • PGVector
  • And others

​
How do I configure LLMs in v0.x?

config = {
    "llm": {
        "provider": "openai",
        "config": {
            "model": "gpt-3.5-turbo",
            "api_key": "your-api-key"
        }
    },
    "version": "v1.0"  # Supported in v0.x
}

m = Memory.from_config(config)

​
Can I use custom prompts in v0.x?

Limited support: 
config = {
    "custom_fact_extraction_prompt": "Your custom prompt here"
    # custom_update_memory_prompt not available in v0.x
}

​
Migration Questions

​
Is migration difficult?

No! Most changes are simple parameter removals: 
# Before (v0.x)
result = m.add("memory", user_id="alice", output_format="v1.1", version="v1.0")

# After (v1.0 Beta)
result = m.add("memory", user_id="alice")

​
Will I lose my data?

No! Your existing memories remain fully compatible with v1.0 Beta.

​
Do I need to re-index my vectors?

No! Existing vector data works with v1.0 Beta without changes.

​
Can I rollback if needed?

Yes! You can always rollback: 
pip install mem0ai==0.1.20  # Last stable v0.x

​
Feature Questions

​
Does v0.x support reranking?

No, reranking is only available in v1.0 Beta: 
# v1.0 Beta only
results = m.search("query", user_id="alice", rerank=True)

​
Can I use advanced filtering in v0.x?

No, only basic key-value filtering: 
# v0.x - basic only
filters = {"category": "food", "user_id": "alice"}

# v1.0 Beta - advanced operators
filters = {
    "AND": [
        {"category": "food"},
        {"score": {"gte": 0.8}}
    ]
}

​
Does v0.x support metadata filtering?

Yes, but basic: 
# Basic metadata filtering
results = m.search(
    "query",
    user_id="alice",
    filters={"category": "work"}
)

​
Performance Questions

​
Is v0.x slower than v1.0 Beta?

Yes, v1.0 Beta includes several performance optimizations:
  • Better async handling
  • Optimized vector operations
  • Improved memory management

​
How do I optimize v0.x performance?

  1. Use async mode when possible
  2. Configure appropriate vector store settings
  3. Use efficient metadata filters
  4. Consider upgrading to v1.0 Beta

​
Can I batch operations in v0.x?

Limited support. Better batch processing available in v1.0 Beta.

​
Troubleshooting

​
Common v0.x Issues

​
1. Inconsistent Response Formats

Problem: Getting different response types Solution: Always use output_format="v1.1"

​
2. Async Mode Not Working

Problem: Async operations failing Solution: Use AsyncMemory class or async_mode=True

​
3. Configuration Errors

Problem: Config not loading properly Solution: Check version parameter and config structure

​
Error Messages

​
”Invalid output format"

# Fix: Use supported format
result = m.add("memory", user_id="alice", output_format="v1.1")

​
"Version not supported"

# Fix: Use supported version
config = {"version": "v1.0"}  # Supported in v0.x

​
"Async mode not available”

# Fix: Use AsyncMemory
from mem0 import AsyncMemory
async_m = AsyncMemory()

​
Getting Help

​
Documentation

​
Community

​
Migration Support

Ready to upgrade? Check out our migration guide to move to v1.0 Beta and access the latest features!