diff --git a/training/configs/qwen2.5_3b_grpo_composite.yaml b/training/configs/qwen2.5_3b_grpo_composite.yaml
index 3c489ffe..ae7fae4b 100644
--- a/training/configs/qwen2.5_3b_grpo_composite.yaml
+++ b/training/configs/qwen2.5_3b_grpo_composite.yaml
@@ -39,8 +39,8 @@ data:
   prompt_key: prompt
   max_prompt_length: 512
   max_response_length: 1024
-  train_batch_size: 64
-  val_batch_size: 64
+  train_batch_size: 32
+  val_batch_size: 32
   return_raw_chat: True
   return_raw_input_ids: True
 
@@ -56,7 +56,7 @@ actor_rollout_ref:
     strategy: fsdp  # This is for backward-compatibility
     ppo_mini_batch_size: 32
     ppo_micro_batch_size: null # will be deprecated, use ppo_micro_batch_size_per_gpu
-    ppo_micro_batch_size_per_gpu: 160
+    ppo_micro_batch_size_per_gpu: 8
     use_dynamic_bsz: False
     ppo_max_token_len_per_gpu: 12288 # n * ${data.max_prompt_length} + ${data.max_response_length}
     grad_clip: 1.0
@@ -107,7 +107,7 @@ actor_rollout_ref:
     free_cache_engine: True
     load_format: dummy_dtensor
     tensor_model_parallel_size: 4
-    max_num_batched_tokens: 8192
+    max_num_batched_tokens: 16384
     max_num_seqs: 1024
     log_prob_micro_batch_size: null # will be deprecated, use log_prob_micro_batch_size_per_gpu
     log_prob_micro_batch_size_per_gpu: 160
@@ -118,7 +118,7 @@ actor_rollout_ref:
     # for hf rollout
     do_sample: True
     use_fire_sampling: False
-    max_model_len: 4096
+    max_model_len: 16384
     # number of responses (i.e. num sample times)
     n: 8 # > 1 for grpo
     val_kwargs:
@@ -178,7 +178,7 @@ critic:
       fsdp_size: -1
   ppo_mini_batch_size: ${actor_rollout_ref.actor.ppo_mini_batch_size}
   ppo_micro_batch_size: null # will be deprecated, use ppo_micro_batch_size_per_gpu
-  ppo_micro_batch_size_per_gpu: null
+  ppo_micro_batch_size_per_gpu: 8
   forward_micro_batch_size: ${critic.ppo_micro_batch_size}
   forward_micro_batch_size_per_gpu: ${critic.ppo_micro_batch_size_per_gpu}
   use_dynamic_bsz: ${actor_rollout_ref.actor.use_dynamic_bsz}