Agent memory systems with LangSmith integration

Tier 1 — battle-tested (21)

• Anthropic Claude (foundation models) Frontier foundation model family (Sonnet 4 / Opus 4.5 / Haiku 4.5)

  Claude family of frontier foundation models — Sonnet 4 (workhorse), Opus 4.5 (frontier capability), Haiku 4.5 (fast/cheap). Frontier model used as substrate by Claude Code, Anthropic Memory tool, Cline, Aider, Continue.dev, Cursor, Goose…

• Mem0 Vector + graph + KV (hybrid)

  Universal memory layer for AI agents. Three concurrent stores (vector + graph + KV); LLM-extracted facts; concurrent retrieval via ThreadPoolExecutor.

• Zep & Graphiti Bi-temporal knowledge graph

  Bi-temporal knowledge graph (event time + ingestion time). Strong on chronological reasoning and contradiction tracking. Graphiti is the open-source core.

• Neo4j Property graph + vectors + Aura Agent

  Property graph DB with Cypher; native vector search; Aura Agent + MCP server for graph-as-memory. $100M GenAI investment in 2025.

• LangGraph (orchestration) Stateful agent graph runtime

  LangChain's stateful graph runtime for agents — explicit state, persistence, time-travel debugging, human-in-the-loop. Memory is one persistence target (LangGraph Persistence cross-listed).

• LanceDB Embedded Arrow-columnar vector DB

  Embedded vector DB (Arrow columnar). RRF reranker. Petabyte-scale on disk.

• Alibaba Qwen 3 family Open-weights frontier model family (Qwen 3 dense + MoE; 0.5B–235B)

  Alibaba Cloud's Qwen 3 family (2025-04) — open-weights frontier models from 0.5B to 235B-A22B MoE. Apache 2.0 (most sizes) — broadest open-weights family. Includes Qwen 3 32B dense, Qwen 3 30B-A3B MoE, Qwen 3 235B-A22B MoE. Qwen-Coder, Q…

• Anthropic Computer Use Beta agent capability for Claude

  Claude 3.5 Sonnet's beta capability for controlling a computer (mouse, keyboard, screen). Frame for browser-agent and OS-agent products built on Claude.

• Anthropic Computer Use Claude desktop-control capability (Anthropic API)

  Anthropic's API-surface capability that lets Claude control a virtual desktop — see screenshots, type, click. Launched 2024-10 as a public beta. Not a finished IDE itself; it's the substrate that other harnesses (and Anthropic's own Clau…

• Browser Use OSS browser-agent library + Cloud (browser-use)

  Python library that turns any LLM into a browser agent — extracts DOM into a structured representation the LLM can act on (rather than pure screenshot). Browser Use Inc. raised $17M seed Q1 2025 (Felicis); cloud product at cloud.browser-…

• CrewAI Multi-agent orchestration

  Multi-agent framework — Crews (sequential) + Flows (event-driven). Open-source core + CrewAI Enterprise. Memory available via Mem0 / Letta adapters (CrewAI Memory cross-listed).

• DeepSeek R1 / V3 family Open-weights reasoning-first frontier model family (R1 / V3 / V3.1)

  DeepSeek (Hangzhou-based, Liang Wenfeng founder) — released DeepSeek-V3 671B MoE (37B active) Dec-2024, then R1 reasoning model Jan-2025 that matched o1 at ~3% of inference cost. Caused 'DeepSeek moment' in markets Jan-27-2025 — NVIDIA d…

• Gemini Enterprise Agent Platform Memory Bank (rebrand of Vertex AI Memory Bank) Long-running agent memory with continuous event-streaming consolidation

  GA at Google Cloud Next 2026 (2026-04-22). Vertex AI rebranded as Gemini Enterprise Agent Platform; Memory Bank now "Agent Platform Memory Bank" with new continuous event-streaming + automated memory generation triggered by configurable …

• Google Gemini Memory Personal Context + Personal Intelligence

  Personal Context setting + Personal Intelligence packing. Memory import from ChatGPT/Claude (Mar 2026).

• Graphiti MCP Server (Zep) Temporal KG with bi-temporal edges

  Tracks how facts change over time; maintains provenance; supports prescribed and learned ontology. Bi-temporal edges (when known vs when valid). Only server that distinguishes "fact learned" from "fact true."

• Grok Memory (xAI) Preference-distillation + user-deletable

  Extracts and persists facts from past conversations; applies them to future responses. Per-memory deletion UI; user can also clear all.

• Mistral Le Chat — Memories Graph-contextual + explicit + implicit

  Le Chat Memories — graph-based, context-aware. Stores user preferences and past decisions; memories saved explicitly (user instruction) or inferred automatically. Coupled to "Work Mode" agentic layer.

• Mistral Vibe (Remote Agents) + Mistral Medium 3.5 Cloud-hosted async coding agents with teleportable session state

  Vibe is Mistral's cloud-hosted coding-agent platform: each session runs in an isolated sandbox with task state, history, and approvals. Spawnable from CLI or Le Chat, with parallel execution. Local CLI sessions can be "teleported" to the…

• OpenAI Agents SDK OpenAI's official agent framework

  OpenAI's official Python framework for agents — handoffs, guardrails, tracing built in. Successor to OpenAI Swarm (experimental); officially supported.

• OpenAI Realtime API Speech-to-speech LLM API (OpenAI Realtime)

  OpenAI's Realtime API for native speech-to-speech multimodal interaction with GPT-4o — launched Oct 2024 (DevDay). Skips the STT → LLM → TTS pipeline by passing speech directly to GPT-4o. The platform substrate that Vapi/Retell/Bland int…

• OpenHands (All Hands AI) OSS autonomous SWE agent platform (formerly OpenDevin)

  Cross-listing of OpenHands at the harness layer. OSS Devin-alternative — sandboxed full SWE agent with browser + shell + editor. Renamed from OpenDevin Mar 2024. ~40k★. Commercial offering: All Hands AI cloud (managed SaaS).