Instructions to use cowWhySo/qwen3_5_2B_voice_assistant-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cowWhySo/qwen3_5_2B_voice_assistant-lora with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="cowWhySo/qwen3_5_2B_voice_assistant-lora")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("cowWhySo/qwen3_5_2B_voice_assistant-lora", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cowWhySo/qwen3_5_2B_voice_assistant-lora with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cowWhySo/qwen3_5_2B_voice_assistant-lora"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cowWhySo/qwen3_5_2B_voice_assistant-lora",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cowWhySo/qwen3_5_2B_voice_assistant-lora

SGLang

How to use cowWhySo/qwen3_5_2B_voice_assistant-lora with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "cowWhySo/qwen3_5_2B_voice_assistant-lora" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cowWhySo/qwen3_5_2B_voice_assistant-lora",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "cowWhySo/qwen3_5_2B_voice_assistant-lora" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cowWhySo/qwen3_5_2B_voice_assistant-lora",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use cowWhySo/qwen3_5_2B_voice_assistant-lora with Docker Model Runner:
```
docker model run hf.co/cowWhySo/qwen3_5_2B_voice_assistant-lora
```

cowWhySo commited on Mar 20

Commit

ce5714d

verified ·

1 Parent(s): 39f488a

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +14 -12

README.md CHANGED Viewed

@@ -2,15 +2,15 @@
 license: apache-2.0
 base_model: unsloth/Qwen3.5-2B
 tags:
-- qwen3.5
-- voice-assistant
-- conversational
-- sft
-- lora
-- unsloth
-- gguf
 language:
-- en
 pipeline_tag: text-generation
 library_name: transformers
 ---
@@ -19,6 +19,8 @@ library_name: transformers
 Fine-tuned [Qwen3.5-2B](https://huggingface.co/unsloth/Qwen3.5-2B) for voice assistant / conversational use.
 Trained on curated, concise datasets — all assistant responses are short and natural-sounding, optimized for spoken output rather than written text.
 ## Training Details
@@ -34,7 +36,7 @@ Trained on curated, concise datasets — all assistant responses are short and n
 | Max sequence length | 1024 |
 | Scheduler | Cosine with 50 warmup steps |
 | Precision | bf16 |
-| Thinking mode | {'Enabled' if ENABLE_THINKING else 'Disabled'} |
 | GPU | NVIDIA L4 (22 GB) |
 | Framework | [Unsloth](https://unsloth.ai) + TRL SFTTrainer |
@@ -49,9 +51,9 @@ bullet points, headings) are excluded. Exact-match deduplication is applied acro
 | [OpenAssistant/oasst_top1_2023-08-25](https://huggingface.co/datasets/OpenAssistant/oasst_top1_2023-08-25) | 2,388 | Real human multi-turn conversations |
 | [HuggingFaceTB/everyday-conversations-llama3.1-2k](https://huggingface.co/datasets/HuggingFaceTB/everyday-conversations-llama3.1-2k) | 1,910 | Greetings, small talk, basic Q&A |
 | [argilla/synthetic-concise-reasoning-sft](https://huggingface.co/datasets/argilla/synthetic-concise-reasoning-sft) | 535 | Short factual reasoning answers |
-| [WizardLM/WizardLM_evol_instruct_70k](https://huggingface.co/datasets/WizardLM/WizardLM_evol_instruct_70k) | 2,000 | Casual single-turn Q&A |
-| **Duplicates removed** | **1,956** | |
-| **Total (after dedup)** | **4,877** | |
 ## Filtering Pipeline (v7)

 license: apache-2.0
 base_model: unsloth/Qwen3.5-2B
 tags:
+  - qwen3.5
+  - voice-assistant
+  - conversational
+  - sft
+  - lora
+  - unsloth
+  - gguf
 language:
+  - en
 pipeline_tag: text-generation
 library_name: transformers
 ---
 Fine-tuned [Qwen3.5-2B](https://huggingface.co/unsloth/Qwen3.5-2B) for voice assistant / conversational use.
+This is designed to be short responses without thinking.
 Trained on curated, concise datasets — all assistant responses are short and natural-sounding, optimized for spoken output rather than written text.
 ## Training Details
 | Max sequence length | 1024 |
 | Scheduler | Cosine with 50 warmup steps |
 | Precision | bf16 |
+| Thinking mode | Disabled |
 | GPU | NVIDIA L4 (22 GB) |
 | Framework | [Unsloth](https://unsloth.ai) + TRL SFTTrainer |
 | [OpenAssistant/oasst_top1_2023-08-25](https://huggingface.co/datasets/OpenAssistant/oasst_top1_2023-08-25) | 2,388 | Real human multi-turn conversations |
 | [HuggingFaceTB/everyday-conversations-llama3.1-2k](https://huggingface.co/datasets/HuggingFaceTB/everyday-conversations-llama3.1-2k) | 1,910 | Greetings, small talk, basic Q&A |
 | [argilla/synthetic-concise-reasoning-sft](https://huggingface.co/datasets/argilla/synthetic-concise-reasoning-sft) | 535 | Short factual reasoning answers |
+| [WizardLM/WizardLM_evol_instruct_70k](https://huggingface.co/datasets/WizardLM/WizardLM_evol_instruct_70k) | 7,000 | Casual single-turn Q&A |
+| **Duplicates removed** | **1,992** | |
+| **Total (after dedup)** | **9,841** | |
 ## Filtering Pipeline (v7)