mcp-titan
3.5
If you are the rightful owner of mcp-titan and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
A neural memory system for LLMs that can learn and predict sequences while maintaining state through a memory vector.
help
Get help about available tools.
init_model
Initialize the Titan Memory model with custom configuration.
forward_pass
Perform a forward pass through the model to get predictions.
train_step
Execute a training step to update the model.
get_memory_state
Get the current memory state and statistics.
manifold_step
Update memory along a manifold direction.
prune_memory
Remove less relevant memories to free up space.
save_checkpoint
Save memory state to a file.
load_checkpoint
Load memory state from a file.
reset_gradients
Reset accumulated gradients to recover from training issues.
