Skip to content

Home

PlugBO: Modular Agentic Bayesian Optimization PlugBO: Modular Agentic Bayesian Optimization

GitHub Blog post

Bayesian optimization (BO) uses a probabilistic surrogate to choose each evaluation, but its overall search strategy is usually fixed in advance. Agentic BO places an LLM agent at the center of the loop: it reviews trial data, surrogate diagnostics, and natural-language context, then decides how to search next. As evidence accumulates, the agent can query or override backend suggestions and revise the surrogate, acquisition function, search bounds, objectives, or constraints1.

Meta's Agentic Bayesian Optimization through Surrogate-Augmented Autoresearch instantiates that idea as Sara, an LLM agent, calling lenz, a BoTorch backend that owns the trial log and posterior computation6. PlugBO keeps that agent and backend, then turns the backend into a plugin surface: surrogate, region, prior, and sampler are slots. Existing and new BO modules wrap as lenz verbs the agent can enable, inspect, or override.

The arrangement is analogous to MCP: Sara is the host, with only bash and read, and lenz is the shared tool surface. A BO method registers extra verbs there the way an MCP server registers tools, so the agent stays fixed while the surrogate, region, prior, or sampler can be swapped in as a plugin.

Note

PlugBO is not an official implementation of Meta's agentic BO paper. What is new here is the plugin protocol, together with new experiments and results. See References for the full source list.

Why PlugBO?

  • Modular and extensible. The surrogate, region, prior, and sampler are slots on a shared BoTorch backend — see Architecture.
  • An LLM agent in a live BO loop. The agent can inspect trials, query the posterior, and reconfigure slots mid-run, while BoTorch retains the trial log and posterior.
  • A shared tool-surface for future BO methods, analogous to MCP. A method registers extra lenz verbs the way an MCP server registers tools — see Plugins.

Quickstart

git clone https://github.com/richardcsuwandi/plugbo.git
cd plugbo
pip install -e .

lenz create --state ./state.json \
  --space '{"x1":{"kind":"range","lower":-5,"upper":10},"x2":{"kind":"range","lower":0,"upper":15}}' \
  --objectives '{"y":"minimize"}' \
  --acqf noisy_logei

lenz suggest --state ./state.json
lenz submit --state ./state.json --config '{"x1":1.0,"x2":2.0}' --metrics '{"y":12.3}'
lenz incumbent --state ./state.json

Continue with the full installation guide and quickstart walkthrough, including running Sara as the agent-in-the-loop.

Modules

Module Role Default Occupant Commands
sara search agent — — sara run
lenz trial log, posterior, acquisition BoTorch loop — create, suggest, submit, incumbent
Surrogate GP fixed Matérn CAKE2 set-surrogate, evolve-kernels, kernel-population
Region search bounds box TuRBO3 set-region, set-bounds, turbo status
Prior belief none πBO4 set-belief
Sampler candidates BoTorch LLAMBO5 set-sampler, llambo sample

  1. Brunzema et al. Agentic Bayesian Optimization through Surrogate-Augmented Autoresearch. arXiv:2608.00316, 2026. ↩

  2. Suwandi et al. Adaptive Kernel Design for Bayesian Optimization Is a Piece of CAKE with LLMs. NeurIPS 2025. ↩

  3. Eriksson et al. Scalable Global Optimization via Local Bayesian Optimization. NeurIPS 2019. ↩

  4. Hvarfner et al. πBO: Augmenting Acquisition Functions with User Beliefs for Bayesian Optimization. ICLR 2022. ↩

  5. Liu et al. Large Language Models to Enhance Bayesian Optimization. ICLR 2024. ↩

  6. Balandat et al. BoTorch: A Framework for Efficient Monte-Carlo Bayesian Optimization. NeurIPS 2020. ↩