Deep Agents Code vs Letta Code

Feature	Deep Agents Code	Letta Code
Tagline	A pre-built coding agent powered by any LLM with tool calling	The memory-first coding agent
Website	https://docs.langchain.com/oss/deepagents/code/overview	https://letta.com
Language	python	typescript
Release Date	October 19, 2025	December 16, 2025
Last Updated	August 19, 2026	August 19, 2026
Provider Type	Independent	Independent
MCP Support	Yes	Yes
Multi-Agent Support	Yes	Yes
Git Integration	basic	basic
Multi-File Editing	Yes	Yes
Auto Test Running	No	Yes
Sandbox / Permissions	Yes	Yes
Image / Multimodal Input	No	Yes
Voice Mode	No	No
IDE Integration	None - Terminal only	None - Terminal only
Hooks / Extensions	Yes	Yes
Web Search	Yes	Yes
Local Model Support	Yes	Yes
Context Window	Model-dependent; compression at threshold	Model-dependent
Terminal Bench	#26 · 66.5% ±3.1 (Terminal-Bench 2.0) 1	#44 · 59.1% ±2.4 (Terminal-Bench 2.0) 2
Supported Models	Any LangChain model: Claude, GPT-5, Gemini, DeepSeek, Ollama, Groq, xAI	Multi-provider with Claude, GPT, Gemini and self-hosted (BYOK).
Supported Subscriptions	ChatGPT	ChatGPT GitHub Copilot Claude Code OpenCode Go / Zen
Pricing Model	BYOK	Freemium
Free Tier	Yes	Yes
Price Range	Free (BYOK)	Free (BYOK) / $20-200/mo
Source Code Availability	Open Source (MIT)	Open Source (Apache-2.0)
Source Code	https://github.com/langchain-ai/deepagents	c
Development State	active	active

¹ Deep Agents Code tested with GPT-5.2-Codex on Terminal-Bench 2.0.

² Letta Code tested with Claude Opus 4.5 on Terminal-Bench 2.0.