# How to use trainer with deepspeed

**URL:** <https://discuss.huggingface.co/t/how-to-use-trainer-with-deepspeed/68872>\
**Category:** Beginners\
**Created:** [January 12, 2024, 2:05am UTC](https://discuss.huggingface.co/t/how-to-use-trainer-with-deepspeed/68872 "2024-01-12T02:05:44Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Kason123](https://avatars.discourse-cdn.com/v4/letter/k/54ee81/32.png) [@Kason123](https://discuss.huggingface.co/u/Kason123)\
**Post date:** [January 12, 2024, 2:05am UTC](https://discuss.huggingface.co/t/how-to-use-trainer-with-deepspeed/68872/1 "2024-01-12T02:05:45Z")

</div>

Hi, I am trying to use deepspeed along with hugginface trainer. For that I simply used the following code. However, this code gives the following error: **RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cuda:1!**. Can someone help me to resolve this issue?

```auto
training_args = TrainingArguments(
    output_dir=outputdir,
    per_device_train_batch_size=script_args.batch_size,
    deepspeed= "deepspeed_config.json",
    bf16=True,
    bf16_full_eval=True,
    gradient_accumulation_steps=script_args.gradient_accumulation_steps,
    learning_rate=script_args.learning_rate,
    logging_steps=script_args.logging_steps,
    num_train_epochs=script_args.num_train_epochs,
    max_steps=script_args.max_steps,
    report_to=script_args.log_with,
    save_steps=script_args.save_steps,
    save_total_limit=script_args.save_total_limit,
    push_to_hub=script_args.push_to_hub,
    hub_model_id=script_args.hub_model_id,
)

trainer = Trainer(
    model,
    training_args,
    train_dataset=tokenized_datasets["train"],
    eval_dataset=tokenized_datasets['eval'],
    tokenizer=tokenizer,
)

```