Dream-RSI SDK · Alpha

Bring your agent. Record its search. Evolve executable exploration policies.

An independent, unofficial Python SDK inspired by Dream-RSI research. Maintained by TheAstrayDev, who is not a Google or Google DeepMind employee. This is a personal research initiative, not a commercial Google development, official SDK or endorsed product.

Python 3.11+ · zero core third-party dependencies · Apache-2.0 · version 0.1.0a4.

Install

python -m pip install dreamrsi==0.1.0a4

from dreamrsi import Budget, DreamRSI

rsi = DreamRSI(
    agent=lambda task: task.upper(),
    evaluator=lambda answer: float(len(answer)),
    budget=Budget(model_calls=4),
)
result = rsi.run_sync("hello")
print(result.best)  # HELLO

This example demonstrates integration, not quality improvement. Stateful adapters support generate/evaluate/refine tasks. Replay uses recorded outcomes without new discovery calls. LLMPolicyDeveloper writes and iteratively rewrites executable source from measured feedback. Version 0.1.0a4 can replay already recorded runs without a new training call, delays holdout collection until a replay-improving candidate exists, and stops repeated policy revisions. The default promotion gate now requires paired raw-quality and probe evidence; applications needing score-only decisions can explicitly use ReplayOnlyGate. These changes reduce avoidable calls, but an end-to-end cost advantage has not been proven. The SDK includes configurable Docker-free policy interpreters, optional process execution, held-out validation, SQLite recovery and reported token/USD accounting.

Tested with Bonsai-27B-Q1_0

A local llama.cpp follow-up scored 5/6 revisions, repaired stopping from 1,000 to 4 replay rounds, promoted a policy and reloaded it for a fresh toy run. Residual error decreased from 1.5 to 0.1875 at six agent calls per policy. This is limited mechanics evidence, not a broad AI benchmark or a reproduction of the paper's performance. Earlier failures and all run denominators remain in the public reports.

• Full documentation and roadmap
• Real-model evidence
• Integration and recovery
• Sandbox configuration
• Issues
• Original Dream-RSI paper

Alpha APIs may change. Generated source runs in a bounded Python-syntax subset, not arbitrary Python. External state isolation and remote cancellation require adapter support. Usage caps depend on accurate provider reports and ceilings. Research-scale performance and broad generalization remain unverified.