Resource hub

Local model orchestration

Coordinate Ollama, open-source models, embeddings, model routing, and private inference flows.

Platform

Hybrid semantic + keyword code search in Atlas (2026)

Atlas, the terminal-native AI coding agent, offers hybrid semantic + keyword code search, fusing results with reciprocal rank fusion. Developers in 2026 can find code by meaning and text, powered by local Ollama

Platform

How to avoid vendor lock-in with an AI coding agent (2026)

SeaShell lets you change the active model and provider mid-session, with favorites and recents, so a price change or a better model is a menu selection rather than a migration.

Platform

Run an AI coding agent with a local model: Ollama and LM Studio in SeaShell (2026)

SeaShell runs against a local model with Ollama or LM Studio. Run seashell models --local to detect your server and register its models, with no per-token API cost.

Platform

Hosted model routing with SeaShell Cloud (2026)

SeaShell Cloud can route model requests through a hosted model router with provider policy, credit checks, and rate limits, while bring-your-own-keys stays the local default.

Browse all SeaShell resource hubs