docs/tests: update default Qwen model to 3.5 0.8B

Author: abetlen

.github/workflows/test.yaml

@@ -8,8 +8,8 @@ on:
       - main
 
 env:
-  REPO_ID: Qwen/Qwen2-0.5B-Instruct-GGUF
-  MODEL_FILE: qwen2-0_5b-instruct-q8_0.gguf
+  REPO_ID: lmstudio-community/Qwen3.5-0.8B-GGUF
+  MODEL_FILE: Qwen3.5-0.8B-Q8_0.gguf
 
 jobs:
   download-model:

README.md

@@ -322,8 +322,8 @@ You'll need to install the `huggingface-hub` package to use this feature (`pip i
 
 ```python
 llm = Llama.from_pretrained(
-    repo_id="Qwen/Qwen2-0.5B-Instruct-GGUF",
-    filename="*q8_0.gguf",
+    repo_id="lmstudio-community/Qwen3.5-0.8B-GGUF",
+    filename="*Q8_0.gguf",
     verbose=False
 )
 ```
@@ -685,7 +685,7 @@ For possible options, see [llama_cpp/llama_chat_format.py](llama_cpp/llama_chat_
 If you have `huggingface-hub` installed, you can also use the `--hf_model_repo_id` flag to load a model from the Hugging Face Hub.
 
 ```bash
-python3 -m llama_cpp.server --hf_model_repo_id Qwen/Qwen2-0.5B-Instruct-GGUF --model '*q8_0.gguf'
+python3 -m llama_cpp.server --hf_model_repo_id lmstudio-community/Qwen3.5-0.8B-GGUF --model '*Q8_0.gguf'
 ```
 
 ### Web Server Features

examples/gradio_chat/local.py

@@ -4,10 +4,10 @@
 import gradio as gr
 
 llama = llama_cpp.Llama.from_pretrained(
-    repo_id="Qwen/Qwen1.5-0.5B-Chat-GGUF",
-    filename="*q8_0.gguf",
+    repo_id="lmstudio-community/Qwen3.5-0.8B-GGUF",
+    filename="*Q8_0.gguf",
     tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "Qwen/Qwen1.5-0.5B"
+        "Qwen/Qwen3.5-0.8B"
     ),
     verbose=False,
 )

examples/hf_pull/main.py

@@ -3,10 +3,10 @@
 
 
 llama = llama_cpp.Llama.from_pretrained(
-    repo_id="Qwen/Qwen1.5-0.5B-Chat-GGUF",
-    filename="*q8_0.gguf",
+    repo_id="lmstudio-community/Qwen3.5-0.8B-GGUF",
+    filename="*Q8_0.gguf",
     tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "Qwen/Qwen1.5-0.5B"
+        "Qwen/Qwen3.5-0.8B"
     ),
     verbose=False,
 )

tests/test_llama.py

@@ -58,8 +58,8 @@ def test_llama_cpp_tokenization():
 
 @pytest.fixture
 def llama_cpp_model_path():
-    repo_id = "Qwen/Qwen2-0.5B-Instruct-GGUF"
-    filename = "qwen2-0_5b-instruct-q8_0.gguf"
+    repo_id = "lmstudio-community/Qwen3.5-0.8B-GGUF"
+    filename = "Qwen3.5-0.8B-Q8_0.gguf"
     model_path = hf_hub_download(repo_id, filename)
     return model_path