Update README.md

README.md

@@ -99,15 +99,10 @@ For more details, please refer to our blog post [Qwen3-Next](https://qwenlm.gith
 
 ## Quickstart
 
-The code for Qwen3-Next has been merged into the main branch of Hugging Face `transformers` since [commit b928235](https://github.com/huggingface/transformers/commit/b9282355bea846b54ed850a066901496b19da654). 
-We strongly recommend using the latest `transformers` main branch.
+The code for Qwen3-Next has been merged into the main branch of Hugging Face `transformers`.
 
 ```shell
-# install from the latest main
 pip install git+https://github.com/huggingface/transformers.git@main
-
-# install the specific commit that includes the qwen3_next support
-# pip install git+https://github.com/huggingface/transformers.git@b9282355bea846b54ed850a066901496b19da654
 ```
 
 With earlier versions, you will encounter the following error:
@@ -125,14 +120,14 @@ model_name = "Qwen/Qwen3-Next-80B-A3B-Instruct"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
-    torch_dtype="auto",
-    device_map="auto"
+    dtype="auto",
+    device_map="auto",
 )
 
 # prepare the model input
 prompt = "Give me a short introduction to large language model."
 messages = [
-    {"role": "user", "content": prompt}
+    {"role": "user", "content": prompt},
 ]
 text = tokenizer.apply_chat_template(
     messages,
@@ -144,7 +139,7 @@ model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
 # conduct text completion
 generated_ids = model.generate(
     **model_inputs,
-    max_new_tokens=16384
+    max_new_tokens=16384,
 )
 output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist() 
 
@@ -161,8 +156,8 @@ print("content:", content)
 > It is recommended to adopt a dedicated inference framework, e.g., SGLang and vLLM, for inference tasks.
 
 > [!Tip]
-> You may observe better efficiency with the `flash-linear-attention` package installed depending on the inference settings. 
-> See [its documentation](https://github.com/fla-org/flash-linear-attention#installation) for detailed instructions and requirements.
+> Depending on the inference settings, you may observe better efficiency with [`flash-linear-attention`](https://github.com/fla-org/flash-linear-attention#installation) and [`causal-conv1d`](https://github.com/Dao-AILab/causal-conv1d).
+> See the above links for detailed instructions and requirements.
 
 
 ## Deployment