Multi-Agent Shared Memory

Agents can share state via shared memory blocks. This allows agents to have a “shared memory”. You can shared blocks between agents by attaching the same block ID to multiple agents.

In the example code below, we create a shared memory block and attach it to a supervisor agent and a worker agent. Because the memory block is shared, when one agent writes to it, the other agent can read the updates immediately.

python

1 from letta_client import Letta
2 
3 client = Letta(base_url="http://127.0.0.1:8283")
4 
5 # create a shared memory block
6 shared_block = client.blocks.create(label="organization", value="Organization: Letta")
7 
8 # create a supervisor agent
9 supervisor_agent = client.agents.create(
10     model="anthropic/claude-3-5-sonnet-20241022",
11     embedding="openai/text-embedding-ada-002",
12     # blocks created for this agentj
13     memory_blocks=[{"label": "persona", "value": "I am a supervisor"}],
14     # pre-existing shared block that is "attached" to this agent
15     block_ids=[shared_block.id],
16 )
17 
18 # create a worker agent
19 worker_agent = client.agents.create(
20     model="anthropic/claude-3-5-sonnet-20241022",
21     embedding="openai/text-embedding-ada-002",
22     # blocks created for this agent
23     memory_blocks=[{"label": "persona", "value": "I am a worker"}],
24     # pre-existing shared block that is "attached" to this agent
25     block_ids=[shared_block.id],
26 )

Memory blocks can also be accessed by other agents, even if not shared. For example, worker agents can write the output of their task to a memory block, which is then read by a supervisor agent. To access the memory blocks of other agents, you can simply use the SDK clients or API to access specific agent’s memory blocks (using the core memory routes).