DeepCode vs letta
DeepCode is an open-source agent for coding. letta is an open-source agent framework for knowledge and memory.
- letta has more GitHub stars (25,466 vs 16,848).
- DeepCode is best for reproduce a research paper as code.
- letta is best for run a coding agent that remembers a project.
- What it is
- DeepCode is an open-source agent for coding.letta is an open-source agent framework for knowledge and memory.
- Best for
- Reproduce a research paper as codeRun a coding agent that remembers a project
- Kind
- AgentAgent
- Pricing
- Open sourceOpen source
- Platforms
- self-hosted, CLICLI, Web, Slack, telegram, discord, self-hosted, API
- Agent type
- Open-source agentAgent framework
- GitHub stars
- 16.8K25.5K
- Last commit
- Nov 2026Oct 2026
- License
- MITApache-2.0
- Open source
- Good for
- Reproduce a research paper as code; Generate a repository from a specification document; Study agentic coding researchRun a coding agent that remembers a project; Add an AI coworker to Slack; Automate recurring work on a schedule
What DeepCode does
Blueprint distillation
Compresses the source document into a plan.
Stateful code memory
Structured index of generated code.
RAG knowledge injection
Pulls in relevant knowledge conditionally.
Closed-loop correction
Detects and fixes errors.
PaperBench evaluation
Reported results on a paper-reproduction benchmark.
What letta does
Stateful agents
Agents carry memory and learn over time.
Letta Code CLI
Open-source harness for coding and assistant work.
Skills and Schedules
Reusable workflows and recurring prompts.
Channels
Slack, Telegram and Discord integrations.
Agent SDK
Embed stateful agents in applications.
Cloud or self-run
Letta Cloud or the App Server.
DeepCode vs letta: questions
Is DeepCode or letta cheaper?
Both are open source. Compare current plans on each official site.
What is the difference between DeepCode and letta?
DeepCode is an open-source agent for coding. letta is an open-source agent framework for knowledge and memory. letta has more GitHub stars (25,466 vs 16,848).

