Support
We're Here to Help
Find instant solutions to common issues or get personalized support from our team
API errors, authentication problems, rate limits, and integration troubleshooting
Login issues, subscription management, billing questions, and account settings
API docs, guides, tutorials, system status, and developer resources
Direct support from our team for complex issues and personalized assistance
Verify your API keys and manage authentication
Check real-time service availability and performance
Browse comprehensive API guides and examples
Send us a detailed message about your issue
Resolve API issues, authentication problems, and Palo memory integration challenges
Palo acts as an intelligent memory layer between your application and any LLM. It automatically captures, processes, and recalls relevant context from conversations, enhancing your chatbot's responses with episodic and semantic memory. Simply integrate our API and Palo handles the memory management transparently.
Episodic memory stores specific conversation events with temporal context ("what happened when"), while semantic memory extracts and organizes conceptual knowledge ("what we know about X"). Palo automatically manages both types.
Free: 100 requests/hour, Pro: 10,000 requests/hour, Enterprise: Custom limits. Most queries complete in <50ms.
Palo supports all major LLM providers (OpenAI, Anthropic, Google, etc.) through our unified API. Configure your preferred providers in Workbench, and Palo will automatically route requests while maintaining consistent memory across all interactions.
Yes! You can set significance thresholds, temporal decay rates, and custom retention policies. Use our /memory/configure endpoint or the Workbench interface to adjust memory parameters.
Enable debug mode with "debug": true in your API calls. The Workbench provides memory inspection tools to visualize retrieved memories.
All memory data is encrypted at rest and in transit. We offer private cloud deployments, GDPR compliance, and granular data retention controls. Memory can be isolated per user, organization, or conversation thread. See our Security page for details.
Use our conversation threading system with unique thread_id parameters. Each thread maintains independent memory while sharing relevant semantic knowledge.
Absolutely! Palo is model-agnostic and works with any text-generation model. Use our Custom Connections feature to integrate proprietary models.
Palo automatically manages capacity using intelligent forgetting algorithms. Less significant memories fade while important information is retained.
Invalid or missing API key in request headers
Bearer sk-your-api-key
Retrieved memory exceeds model context window
max_memory_tokens parameter or use memory summarization
Memory bleeding between conversation threads
thread_id for each conversation and use proper isolation settings
Exceeded rate limits for memory operations
Memory queries taking longer than expected
Reached storage limits for your plan
Get up and running with Palo in under 10 minutes
Python, JavaScript, and REST API references
Real-world implementations and best practices
Manage your Mpalo account, memory storage subscriptions, and billing preferences
Visit your Billing Dashboard to compare plans and upgrade instantly. Higher tiers include more memory storage, faster retrieval, advanced forgetting controls, and priority support. Changes take effect immediately.
Yes! The Workbench dashboard provides real-time usage analytics including memory storage consumed, API calls made, retrieval latency, and cost breakdowns. Set up usage alerts to avoid overages.
Your memory data is preserved but access may be limited to your new plan's storage quota. Palo's intelligent forgetting will prioritize the most significant memories. We recommend exporting important memories before downgrading.
Account deletion is permanent and irreversible. Go to Workbench Settings → Account → Delete Account. All memory data, API keys, and billing history will be permanently removed within 30 days.
Yes! Enterprise accounts include shared memory pools, team management, usage analytics, custom retention policies, dedicated support, and volume discounts. Contact our sales team for custom pricing and implementation.
Memory storage is billed monthly based on your peak usage. API calls are metered separately. We offer predictable pricing tiers and usage-based scaling. Check your current usage and costs in the Workbench billing section.
Yes, you can pause your subscription for up to 90 days. Your memory data is preserved but API access is suspended. Reactivation restores full functionality. This is useful for seasonal applications or development breaks.
We accept all major credit cards, PayPal, and ACH transfers for enterprise accounts. Payments are processed securely through Stripe. Enterprise customers can also arrange annual billing and custom payment terms.
Can't access your Mpalo account or Workbench
Lost access to 2FA device or backup codes
Account suspended due to suspicious activity
Credit card declined or payment processing error
Higher than expected usage or billing amounts
Need detailed billing breakdown or tax documents
Comprehensive guides, API documentation, and developer resources for Palo AI Memory
Memory systems for autonomous robots and embodied AI
Pre-built memory configurations for common use cases
Connect proprietary models and custom endpoints
Deep dive into memory engine architecture and tuning
Complete API reference, SDKs, and integration examples
Real-time service status and performance metrics
Interactive playground and account management dashboard
Can't find what you're looking for? Our support team is here to help with personalized assistance.
Response time: 24-48 hours (business days) • Email: support@mpalo.com