World Model Optimizer
wmo turns agent traces you already collect into continuous improvement. Start with a model
endpoint at frontier quality with 40%+ lower cost. Keep improving it with world model simulations,
meta-harness optimization, and model distillation.
🌐 Platform |
📚 Docs |
Discord
Getting started
- Register your providers.
- Tune a router on your OTel traces.
- Serve it.
See what it bought you against the model you were using before:
Distill your own small model into the pool with wmo optimize model,
serve a single model with no routing via wmo optimize route pin, or build an optimized harness
for your agent with wmo optimize harness.
Hosted platform
Create an account at platform.experientiallabs.ai, then
authenticate the CLI:
Copy an agent ID from the platform and run its current champion harness:
E2B backend
Hosted agents already run in platform-managed E2B sandboxes. To evaluate a local optimization in
E2B, install the extra and provide an E2B key:
Use a world model as an API
world-model-optimizer includes world models that can be used to simulate your agent environment
for testing and optimization.
Or over HTTP (same code path), namespaced by model name: GET /world_models, then POST /world_models/{name}/sessions and POST /world_models/{name}/sessions/{id}/step.
Run after platform login
After wmo login, the same wmo run command can open a hosted world model or run an agent's
current champion harness in E2B. The platform manages model and sandbox credentials, so hosted
runs do not need local API keys.
Workspace upload is opt-in with -u: WMO live-syncs changes and preserves concurrent local edits.
Long-running agents can detach, continue in the platform, and be messaged or reattached later.
Runtime agents and optimizers in E2B sandboxes
WMO can run the real pi worker inside isolated
E2B sandboxes while the world model supplies the environment. Optimization and
evaluation rollouts run in parallel, and model credentials stay outside the sandbox.
The optimizer can change prompts, tools, policies, skills, and runtime code. Every candidate is
measured against the same simulated tasks, and only changes that pass the evaluation gates become
the new versioned champion harness.
Development
Managed with uv; linting/formatting with ruff; type checking with ty. Conventions live in AGENTS.md.
Usage telemetry
wmo uses anonymous usage telemetry to track the volume of usage.
Telemetry is strictly metadata. It never includes prompts, traces, actions, observations, file
paths,
model names, provider credentials, or raw user content.
Telemetry is enabled by default. To opt out for a project:
This writes .wmo/settings.toml. You can re-enable it with uv run wmo config telemetry enable,
check the current setting with uv run wmo config telemetry status, or disable it for a process
with DO_NOT_TRACK=1 or WMO_TELEMETRY=0.