OpenRouter Model Stacks for AI Assistants
A model stack is a curated set of AI models that your OpenClaw assistant routes across intelligently. Instead of locking you to one model, the stack hands every message to OpenRouter’s auto-router, which picks the best model from the set for that specific request, a hard coding question might go to the strongest model, a quick reply to a faster, cheaper one. Each stack also has a built-in cost-vs-quality lean, so the routing matches the stack’s intent.
Stacks at a glance
Free models only. Explore without spending a credit.
Efficient paid models. Quality stays high, spend stays low.
Balanced quality and speed for everyday work.
The best models available. For when quality is everything.
How we choose models
Every stack has different criteria. We cross-reference PinchBench (real-world task success rates and cost-per-run value scores) with OpenRouter (live pricing, throughput, availability) before making any changes.
| Stack | Capability bar | Cost bar |
|---|---|---|
| Wanderer | Solid task completion | Must be :free on OpenRouter |
| Hustler | High success rate, strong value score | Lowest cost-per-run with acceptable quality |
| Professional | High success rate, fast throughput | Mid cost, latency matters here |
| Operator | Best available, leads on reasoning | Cost is secondary |
Model details and benchmarks
CX23-safe free coding model with tool support. New to the stack, so availability and real-world quality are being monitored.
Best-scoring free model on PinchBench. Single endpoint with variable uptime, so it backs up rather than leads.
Swapped in from the 31B variant (Model Scout, Jul 2026) — 2 providers instead of 1, best value score on PinchBench. Also handles image messages for this stack.
Optimised for ultra-low latency and cost efficiency. Built-in reasoning via API.
New paid saver candidate with three providers, configurable reasoning, tool use, and a safe 262K/131K context-output profile.
Reliable OpenAI fallback. Consistent availability and broad task coverage.
OpenAI's cost-sensitive GPT-5.6 tier. Five live endpoints and a safe 1M/128K context-output profile.
Near-frontier intelligence at low latency. Matches Sonnet 4 on reasoning, coding, and computer use.
OpenAI's balanced GPT-5.6 tier for everyday coding, reasoning, and agentic work.
Excellent OpenClaw benchmark performance and value. Single-provider availability is covered by the stack's outer fallback.
Frontier performance across coding, agents, and professional work. Leads on instruction-following and complex reasoning.
PinchBench top-tier model with seven endpoints. Every live provider reports a safe 128K max output on its 1M context.
OpenAI's current flagship for complex professional work, coding, and agentic reasoning.
Added Jul 2026 (Model Scout) — xAI's first entry in our stacks. Coding/agent-focused, 4 xAI-operated endpoints.
How auto-routing works
For each message, OpenRouter’s auto-router looks at the models in your stack and picks the best fit for that request, weighing capability against cost using the stack’s lean (Operator leans hard toward quality; Wanderer and Hustler lean toward savings; Professional sits in the middle). Harder tasks pull a stronger model; simple ones get a faster, cheaper one. You don’t configure anything, it happens per message, automatically, within the set you’ve chosen. If the router itself ever has a hiccup, a reliable safety model in the stack catches it, so there’s no downtime.
Credit usage
Costs vary by model and message length. Rough per-message estimates:
| Stack | Typical cost per message |
|---|---|
| Wanderer | $0 |
| Hustler | $0.001–$0.005 |
| Professional | $0.003–$0.015 |
| Operator | $0.015–$0.08 |
Check your Dashboard usage tab to monitor actual spend.
Switching stacks
Open your Dashboard, go to Settings, and select a new stack. The change applies immediately, no redeploy needed. You can also ask your bot to use any individual model for a specific conversation without changing your stack.
Model compatibility
Stack models are tested for compatibility with the CX23 VPS before any changes are made. If you use the custom model picker, be aware that some newer 1M context models have a token specification issue that causes immediate overflow errors. See VPS specs and model limits for the full details.