run_qwen3-235b_megatron_96gb.sh

#!/usr/bin/env bash
set -xeuo pipefail

## !!!!!!!important!!!!!!
## set the following environment variables on all your nodes
# env_vars:
#   CUDA_DEVICE_MAX_CONNECTIONS: "1"
#   NCCL_NVLS_ENABLE: "0"
#   VLLM_USE_V1: 1
# install mbridge=0.1.13 on all your node with the following command: 
# pip3 install git+https://github.com/ISEEKYAN/mbridge

SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
[ -f "${SCRIPT_DIR}/env.sh" ] && source "${SCRIPT_DIR}/env.sh"

adv_estimator=grpo

use_kl_in_reward=False
kl_coef=0.0
use_kl_loss=True
kl_loss_coef=0.001

clip_ratio_low=0.2
clip_ratio_high=0.28

max_prompt_length=$((1024 * 2))
max_response_length=$((1204 * 8))
enable_overlong_buffer=True
overlong_buffer_len=$((1024 * 1))
overlong_penalty_factor=1.0

loss_agg_mode="token-mean"

train_prompt_bsz=${TRAIN_BS:-32}
n_resp_per_prompt=8
train_prompt_mini_bsz=16

# minimum nodes need for qwen3-235B-A22B
NNODES=${NNODES:-4}
# Paths

RAY_DATA_HOME=${RAY_DATA_HOME:-"${HOME}/verl"}

MODEL_PATH=$RAY_DATA_HOME/models/Qwen3-235B-A22B

TRAIN_FILE=$RAY_DATA_HOME/dataset/dapo-math-17k.parquet
TEST_FILE=$RAY_DATA_HOME/dataset/aime-2024.parquet

# Algorithm
temperature=1.0
top_p=1.0
top_k=-1 # 0 for HF rollout, -1 for vLLM rollout
val_top_p=0.7
# Performance Related Parameter
use_dynamic_bsz=True
actor_ppo_max_token_len=$(((max_prompt_length + max_response_length) * 10 / 10))
infer_ppo_max_token_len=$(((max_prompt_length + max_response_length) * 1))
offload=True
OPTIM_OFFLOAD=${OPTIM_OFFLOAD:-True}
gen_tp=8
train_tp=${TP:-4}
train_pp=${PP:-8}

EP=${EP:-4}
ETP=1
CP=1
optimizer_offload_fraction=${OFFLOAD_FRACTION:-1.}
last_layer=${LAST_LAYER:-10}

project_name='verl-qwen3'
exp_name="235B-${NNODES}-pp${train_pp}-tp${train_tp}-ep${EP}-actor-length${actor_ppo_max_token_len}"
CKPTS_DIR=$RAY_DATA_HOME/ckpt/${project_name}/${exp_name}

# TODO: support cuda graph for rollout by setting the following config
    # actor_rollout_ref.rollout.cudagraph_capture_sizes=[1,2,4,8,16,32]
    # actor_rollout_ref.rollout.enforce_eager=False

python3 -m verl.trainer.main_ppo \
    --config-path=config \
    --config-name='ppo_megatron_trainer.yaml' \
    data.train_files="${TRAIN_FILE}" \
    data.val_files="${TEST_FILE}" \
    data.prompt_key=prompt \
    data.truncation='left' \
    data.max_prompt_length=${max_prompt_length} \
    data.max_response_length=${max_response_length} \
    data.train_batch_size=${train_prompt_bsz} \
    actor_rollout_ref.rollout.n=${n_resp_per_prompt} \
    actor_rollout_ref.rollout.name=vllm \
    actor_rollout_ref.rollout.enforce_eager=True \
    actor_rollout_ref.rollout.free_cache_engine=True \
    algorithm.adv_estimator=${adv_estimator} \
    algorithm.use_kl_in_reward=${use_kl_in_reward} \
    algorithm.kl_ctrl.kl_coef=${kl_coef} \
    actor_rollout_ref.model.use_fused_kernels=True \
    actor_rollout_ref.actor.megatron.use_mbridge=True \
    actor_rollout_ref.actor.use_kl_loss=${use_kl_loss} \
    actor_rollout_ref.actor.kl_loss_coef=${kl_loss_coef} \
    actor_rollout_ref.actor.clip_ratio_low=${clip_ratio_low} \
    actor_rollout_ref.actor.clip_ratio_high=${clip_ratio_high} \
    actor_rollout_ref.actor.clip_ratio_c=10.0 \
    actor_rollout_ref.actor.ppo_micro_batch_size_per_gpu=2 \
    actor_rollout_ref.ref.log_prob_micro_batch_size_per_gpu=4 \
    actor_rollout_ref.rollout.log_prob_micro_batch_size_per_gpu=4 \
    actor_rollout_ref.actor.use_dynamic_bsz=${use_dynamic_bsz} \
    actor_rollout_ref.ref.log_prob_use_dynamic_bsz=${use_dynamic_bsz} \
    actor_rollout_ref.rollout.log_prob_use_dynamic_bsz=${use_dynamic_bsz} \
    actor_rollout_ref.actor.ppo_max_token_len_per_gpu=${actor_ppo_max_token_len} \
    actor_rollout_ref.ref.log_prob_max_token_len_per_gpu=${infer_ppo_max_token_len} \
    actor_rollout_ref.rollout.log_prob_max_token_len_per_gpu=${infer_ppo_max_token_len} \
    actor_rollout_ref.model.path="${MODEL_PATH}" \
    actor_rollout_ref.actor.optim.lr=1e-6 \
    actor_rollout_ref.actor.optim.lr_warmup_steps=10 \
    actor_rollout_ref.actor.optim.weight_decay=0.1 \
    +actor_rollout_ref.actor.optim.override_optimizer_config.optimizer_offload_fraction=${optimizer_offload_fraction} \
    +actor_rollout_ref.actor.optim.override_optimizer_config.overlap_cpu_optimizer_d2h_h2d=True \
    +actor_rollout_ref.actor.optim.override_optimizer_config.use_precision_aware_optimizer=True \
    +actor_rollout_ref.actor.optim.override_optimizer_config.optimizer_cpu_offload=True \
    actor_rollout_ref.actor.ppo_mini_batch_size=${train_prompt_mini_bsz} \
    actor_rollout_ref.actor.megatron.param_offload=${offload} \
    actor_rollout_ref.actor.megatron.optimizer_offload=${OPTIM_OFFLOAD} \
    actor_rollout_ref.actor.megatron.grad_offload=${offload} \
    actor_rollout_ref.actor.megatron.pipeline_model_parallel_size=${train_pp} \
    actor_rollout_ref.actor.megatron.tensor_model_parallel_size=${train_tp} \
    actor_rollout_ref.actor.megatron.expert_model_parallel_size=$EP \
    actor_rollout_ref.actor.megatron.expert_tensor_parallel_size=$ETP \
    actor_rollout_ref.actor.megatron.context_parallel_size=${CP} \
    actor_rollout_ref.actor.entropy_coeff=0 \
    actor_rollout_ref.actor.optim.clip_grad=1.0 \
    actor_rollout_ref.actor.loss_agg_mode=${loss_agg_mode} \
    actor_rollout_ref.rollout.gpu_memory_utilization=0.85 \
    actor_rollout_ref.rollout.tensor_model_parallel_size=${gen_tp} \
    actor_rollout_ref.rollout.enable_chunked_prefill=True \
    actor_rollout_ref.rollout.max_num_batched_tokens=$((max_prompt_length + max_response_length)) \
    actor_rollout_ref.rollout.temperature=${temperature} \
    actor_rollout_ref.rollout.top_p=${top_p} \
    actor_rollout_ref.rollout.top_k=${top_k} \
    actor_rollout_ref.nccl_timeout=1200 \
    actor_rollout_ref.rollout.val_kwargs.temperature=${temperature} \
    actor_rollout_ref.rollout.val_kwargs.top_p=${val_top_p} \
    actor_rollout_ref.rollout.val_kwargs.top_k=${top_k} \
    actor_rollout_ref.rollout.val_kwargs.do_sample=True \
    actor_rollout_ref.rollout.val_kwargs.n=1 \
    actor_rollout_ref.ref.megatron.pipeline_model_parallel_size=${train_pp} \
    actor_rollout_ref.ref.megatron.tensor_model_parallel_size=${train_tp} \
    actor_rollout_ref.ref.megatron.expert_model_parallel_size=$EP \
    actor_rollout_ref.ref.megatron.expert_tensor_parallel_size=$ETP \
    actor_rollout_ref.ref.megatron.context_parallel_size=${CP} \
    actor_rollout_ref.ref.megatron.param_offload=${offload} \
    +actor_rollout_ref.actor.megatron.override_transformer_config.apply_rope_fusion=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.masked_softmax_fusion=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.bias_activation_fusion=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.bias_dropout_fusion=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.gradient_accumulation_fusion=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.deallocate_pipeline_outputs=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.persist_layer_norm=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.moe_grouped_gemm=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.moe_permute_fusion=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.moe_token_dispatcher_type="flex" \
    +actor_rollout_ref.actor.megatron.override_transformer_config.moe_router_dtype=fp32 \
    +actor_rollout_ref.actor.megatron.override_transformer_config.moe_enable_deepep=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.account_for_loss_in_pipeline_split=True \
    +actor_rollout_ref.actor.megatron.override_transformer_config.account_for_embedding_in_pipeline_split=True \
    reward.reward_manager.name=dapo \
    +reward.reward_kwargs.overlong_buffer_cfg.enable=${enable_overlong_buffer} \
    +reward.reward_kwargs.overlong_buffer_cfg.len=${overlong_buffer_len} \
    +reward.reward_kwargs.overlong_buffer_cfg.penalty_factor=${overlong_penalty_factor} \
    +reward.reward_kwargs.overlong_buffer_cfg.log=False \
    +reward.reward_kwargs.max_resp_len=${max_response_length} \
    trainer.logger=['console','wandb'] \
    trainer.project_name="${project_name}" \
    trainer.experiment_name="${exp_name}" \
    trainer.n_gpus_per_node=8 \
    trainer.nnodes="${NNODES}" \
    trainer.val_before_train=False \
    trainer.test_freq=10 \
    trainer.save_freq=100 \
    trainer.total_epochs=10 \
    trainer.default_local_dir="${CKPTS_DIR}" \
    trainer.resume_mode=auto \
    trainer.log_val_generations=10