94 lines
2.3 KiB
Markdown
94 lines
2.3 KiB
Markdown
|
---
|
||
|
|
license: apache-2.0
|
||
|
|
datasets:
|
||
|
|
- NousResearch/Hermes-3-Dataset
|
||
|
|
- HuggingFaceTB/everyday-conversations-llama3.1-2k
|
||
|
|
base_model:
|
||
|
|
- Qwen/Qwen3-4B
|
||
|
|
---
|
||
|
|
|
||
|
|
This Qwen 3 4B model was fine-tuned on the Hermes 3 dataset to enhance its general chatting capabilities while retaining Qwen's Reasoning capabilities.
|
||
|
|
|
||
|
|
## transformers
|
||
|
|
As the qwen team suggested to use
|
||
|
|
```python
|
||
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||
|
|
|
||
|
|
model_name = "ertghiu256/Qwen3-Hermes-4b"
|
||
|
|
|
||
|
|
# load the tokenizer and the model
|
||
|
|
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
||
|
|
model = AutoModelForCausalLM.from_pretrained(
|
||
|
|
model_name,
|
||
|
|
torch_dtype="auto",
|
||
|
|
device_map="auto"
|
||
|
|
)
|
||
|
|
|
||
|
|
# prepare the model input
|
||
|
|
prompt = "Give me a short introduction to large language model."
|
||
|
|
messages = [
|
||
|
|
{"role": "user", "content": prompt}
|
||
|
|
]
|
||
|
|
text = tokenizer.apply_chat_template(
|
||
|
|
messages,
|
||
|
|
tokenize=False,
|
||
|
|
add_generation_prompt=True,
|
||
|
|
enable_thinking=True # Switches between thinking and non-thinking modes. Default is True.
|
||
|
|
)
|
||
|
|
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
|
||
|
|
|
||
|
|
# conduct text completion
|
||
|
|
generated_ids = model.generate(
|
||
|
|
**model_inputs,
|
||
|
|
max_new_tokens=32768
|
||
|
|
)
|
||
|
|
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
|
||
|
|
|
||
|
|
# parsing thinking content
|
||
|
|
try:
|
||
|
|
# rindex finding 151668 (</think>)
|
||
|
|
index = len(output_ids) - output_ids[::-1].index(151668)
|
||
|
|
except ValueError:
|
||
|
|
index = 0
|
||
|
|
|
||
|
|
thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
|
||
|
|
content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")
|
||
|
|
|
||
|
|
print("thinking content:", thinking_content)
|
||
|
|
print("content:", content)
|
||
|
|
```
|
||
|
|
|
||
|
|
## vllm
|
||
|
|
Run this command
|
||
|
|
```bash
|
||
|
|
vllm serve ertghiu256/Qwen3-Hermes-4b --enable-reasoning --reasoning-parser deepseek_r1
|
||
|
|
```
|
||
|
|
|
||
|
|
## Sglang
|
||
|
|
Run this command
|
||
|
|
```bash
|
||
|
|
python -m sglang.launch_server --model-path ertghiu256/Qwen3-Hermes-4b --reasoning-parser deepseek-r1
|
||
|
|
```
|
||
|
|
|
||
|
|
## llama.cpp
|
||
|
|
Run this command
|
||
|
|
```bash
|
||
|
|
llama-server --hf-repo ertghiu256/Qwen3-Hermes-4b
|
||
|
|
```
|
||
|
|
or
|
||
|
|
```bash
|
||
|
|
llama-cli --hf ertghiu256/Qwen3-Hermes-4b
|
||
|
|
```
|
||
|
|
|
||
|
|
## ollama
|
||
|
|
Run this command
|
||
|
|
```bash
|
||
|
|
ollama run hf.co/ertghiu256/Qwen3-Hermes-4b:Q4_K_M
|
||
|
|
```
|
||
|
|
|
||
|
|
## lm studio
|
||
|
|
Search
|
||
|
|
```
|
||
|
|
ertghiu256/Qwen3-Hermes-4b
|
||
|
|
```
|
||
|
|
in the lm studio model search list then download
|