tinker Backend

import hydra
from omegaconf import DictConfig

from examples.math_tinker.math_agent_with_fewshot import MathAgentWithFewshot
from examples.math_tinker.math_reward import math_reward_fn
from rllm.data.dataset import DatasetRegistry
from rllm.environments.base.single_turn_env import SingleTurnEnvironment
from rllm.trainer import AgentTrainer

@hydra.main(
    version_base=None,
    config_path="../../rllm/trainer/config",
    config_name="tinker_rl_trainer"
)
def main(config: DictConfig):
    # Load datasets
    train_dataset = DatasetRegistry.load_dataset("gsm8k", "train")
    test_dataset = DatasetRegistry.load_dataset("math500", "test")

    # Create trainer with tinker backend
    trainer = AgentTrainer(
        config=config,
        agent_class=MathAgentWithFewshot,
        env_class=SingleTurnEnvironment,
        agent_args={"use_fewshot": True},
        env_args={"reward_fn": math_reward_fn},
        train_dataset=train_dataset,
        val_dataset=test_dataset,
        backend="tinker",  # Specify tinker backend
    )

    # Train
    trainer.train()

if __name__ == "__main__":
    main()

import hydra
from omegaconf import DictConfig

from examples.solver_judge_tinker.solver_judge_flow import SolverJudgeFlow
from rllm.data.dataset import DatasetRegistry
from rllm.trainer import WorkflowTrainer

@hydra.main(
    version_base=None,
    config_path="../../rllm/trainer/config",
    config_name="tinker_rl_trainer"
)
def main(config: DictConfig):
    train_dataset = DatasetRegistry.load_dataset("countdown", "train")
    test_dataset = DatasetRegistry.load_dataset("countdown", "test")

    trainer = WorkflowTrainer(
        config=config,
        workflow_class=SolverJudgeFlow,
        workflow_args={},
        train_dataset=train_dataset,
        val_dataset=test_dataset,
        backend="tinker",
    )

    trainer.train()

if __name__ == "__main__":
    main()

# Model
model:
  name: "Qwen/Qwen3-8B"
  lora_rank: 32
  train_unembed: true
  train_attn: true
  train_mlp: true

# Training
training:
  group_size: 16
  val_group_size: 1
  learning_rate: 2e-5
  max_length: 32768

# Sampling
sampling:
  temperature: 1.0
  top_p: 1.0

# Algorithm
algorithm:
  adv_estimator: grpo
  gamma: 1.0
  lam: 0.95
  norm_adv_by_std_in_grpo: false
  grouping_level: 'trajectory'

# Data
data:
  train_batch_size: 64
  val_batch_size: 32
  max_prompt_length: 2048
  max_response_length: 2048

# Trainer
trainer:
  total_epochs: 10
  test_freq: 5
  save_freq: 20
  logger: ['console', 'wandb']
  project_name: 'math-rl'
  experiment_name: 'qwen3-8b-gsm8k'
  default_local_dir: '/tmp/rllm-tinker-checkpoints'

# Agent
agent:
  max_steps: 1  # Single-turn
  agent_args: {}

# Environment
env:
  env_args: {}

# Rollout Engine
rollout_engine:
  reasoning_effort: "medium"
  accumulate_reasoning: false
  disable_thinking: false