Find the shortest route through your cities — one of computing's oldest hard problems, solved live in your browser with 17 algorithms, from exact search to genetic algorithms and swarm intelligence.
New to the problem? Read the TSP overview — history, real-world uses, and further reading.
Drag a .tsp file here
or
Drag a .opt.tour file here
or
optional reference tour
Select a solver above to see its options.
Solve TSP problems directly from an AI agent in the browser — no server, no API key, just a page an agent can call.
A REST API for solving TSP problems programmatically, with OpenAPI docs and rate limiting built in.
A command-line binary for scripts and pipelines, plus a graphical desktop app built with Qt.
Every solver is compiled to WebAssembly and runs client-side at near-native speed — no install, nothing sent to a server.
Read how each solver works, with pseudocode and references:
Explores the full search space to guarantee the shortest possible tour — only practical for a few dozen cities.
Builds a tour in a single pass with a greedy rule or geometric heuristic — fast, but rarely optimal on its own.
Starts from a complete tour and keeps swapping edges until no swap can shorten it any further.
Searches broadly with randomness or swarm behavior to escape the local optima that trap local search.