AI Agent Frameworks & MCP — Market

Verified claims and product-axis read for AI Agent Frameworks & MCP. Every fact below is sourced; every product judgment traces back to underlying signals.

Verified facts

• LangSmith offers a free tier with 5K traces/month and Plus at $39/seat/month, Enterprise via sales-contact ↗ (financial)
• Median GitHub issue resolution time for LlamaIndex is under 7 days vs over 20 days for LangChain ↗ (other)
• Arize, Langfuse, and Helicone are positioned as third-party observability alternatives to LangSmith ↗ (other)
• LlamaIndex has under 30 full-time employees as of 2025 ↗ (other)
• LangChain core team is approximately 50-70 employees as of 2025 ↗ (other)
• deepset has approximately 60-80 employees primarily in Berlin and remote EU ↗ (other)
• LangChain has an OpenAI integration that is downloaded as the langchain-openai sub-package over 8M times/month ↗ (other)
• LangChain's revenue was reported in industry press to be approximately $12-16M ARR in early 2025 ↗ (financial)
• deepset positions Haystack as 'production-grade' with a public reliability SLA for deepset Cloud ↗ (other)
• LlamaIndex launched LlamaParse as a separate paid product in 2024 with usage-based PDF parsing pricing ↗ _(historical_event)_

Top products (engine read)

Local / Sovereign AI Agent Framework (GAIA, Kestrel, CODEC) — on-device / offline-capable agent runtime

Opportunity: Four independent OSS frameworks in one window (GAIA, Kestrel, Esp-Claw, CODEC) all pitching 'runs locally / sovereign / on edge' — clear demand for a non-cloud agent stack that LangChain/CrewAI/LlamaIndex (all hosted-leaning) don't address.

Sovereign / local-runtime agent frameworks are a true white-space: no tracked incumbent is committing, yet 4 distinct community projects launched in-window. Likely an opportunity zone for the next 12 months, especially for EU + IoT buyers.

Browser / OS Automation Frameworks for AI Agents (Libretto, SoMatic) — deterministic browser & vision-based RPA layer for agents

Opportunity: Two distinct show-HN launches both flagging the same gap: LLM agents can 'see' but can't deterministically act on browser/OS. This is the wedge below the orchestration layer — and neither LangChain/CrewAI/LlamaIndex own it.

Browser/OS determinism is the bottleneck for shipping computer-use agents to production. Likely a hot acquisition target for an orchestrator vendor (LangChain or OpenAI) in the next 6–12 months.

Agent Memory Framework (DMF) — deterministic / long-term memory layer for agents

Opportunity: Post 36163815 directly names 'better memory systems' as the #1 missing feature; post 7362f1bd surveys 30+ frameworks on 'context rot, memory and tools'; DMF launches as a dedicated answer.

Memory is the most-cited gap in current agent frameworks. Standalone memory primitives (DMF-style) and integration into LangGraph-style orchestrators are both viable paths — currently no clear winner.

TypeScript Agent Frameworks (Better Agent, reactive/non-blocking TS frameworks) — type-safe end-to-end TS agent runtime

Opportunity: 5 distinct TS-agent posts in one window — clear pull from JS/Next.js devs locked out of the Python-first LangChain/LlamaIndex/CrewAI/Haystack stack. The incumbents (all Python-primary) are not directly addressing this audience.

TypeScript is a structurally underserved gap in the agent framework topic. Whoever wins it (a la 'LangChain for Node') captures full-stack web devs that the Python incumbents can't easily reach.

See the Products and Strategy modules for the full product list and forward-looking judgment.