← Back to home
Comparison · ai-assistants

Semantic Kernel vs Exa

Side-by-side trajectory, velocity, and editorial themes.

S
Semantic Kernel
AI-ASSISTANTS
3.8

Semantic Kernel ships steady .NET/Python point releases while pointing users to its successor framework.

◆ Current state

Microsoft's Semantic Kernel releases as parallel per-language package trains (.NET and Python), each a mix of dependency bumps, security hardening, and occasional real capability work. Recent notes add HTTP-redirect disabling and file-path validation hardening on .NET, OpenAPI parsing and server-URL validation changes, and Assistant-agent function-choice support on Python. Several release notes carry a documented callout naming the Microsoft Agent Framework as SK's successor.

◆ Where it's heading

The engineering signal is maintenance-plus: dependency currency, security tightening, and API refinement rather than large new capability surfaces. The more consequential thread is positional — SK is steering developers toward the Microsoft Agent Framework, which frames this train as stabilization of an established codebase rather than expansion.

◆ Prediction

Expect continued incremental point releases focused on security, dependency updates, and OpenAPI/agent API polish, alongside more explicit migration signposting toward the Agent Framework.

E
Exa
AI-ASSISTANTS
6.3

Exa is pushing past search into autonomous web-research agents.

◆ Current state

Exa has moved beyond its search-and-retrieval API into agentic territory. The headline change is Exa Agent — a research agent built on Exa's index and reachable via API — now joined by MCP availability for Agent and Connect. The underlying search product keeps maturing in parallel: auto-routing, people and company search, markdown-native content, and instant results.

◆ Where it's heading

The arc runs from primitives to products: a fast index, then specialized verticals (people, companies), now an agent that composes them into end-to-end research. Bringing Agent and Connect to MCP signals Exa wants to be a retrieval backend inside other agent stacks, not just a standalone API.

◆ Prediction

Expect Exa to deepen the agent layer — structured research outputs and monitoring already appear in the changelog — and to lean on MCP distribution to embed inside third-party agents rather than compete for end users directly.

See more alternatives to Semantic Kernel
See more alternatives to Exa