Qwen3 8B Esper3

🚀 Esper 3: A Coding and Reasoning Specialist

Esper 3 is a specialized model for coding, architecture, and DevOps reasoning, built upon the Qwen 3 framework. It offers high - performance reasoning capabilities in various technical domains and is suitable for a wide range of users, from developers to architects.

Support our open - source dataset and model releases!

Available Versions

• Qwen3 - 4B
• Qwen3 - 8B
• Qwen3 - 14B

✨ Features

• Fine - Tuned on Specialized Data: Esper 3 is finetuned on DevOps and architecture reasoning and code reasoning data generated with Deepseek R1.
• Improved Reasoning: It has enhanced general and creative reasoning to boost problem - solving and general chat performance.
• Small Model Sizes: Allows for local desktop and mobile deployment, along with super - fast server inference.

📦 Model Information

💻 Usage Examples

Basic Usage

Esper 3 uses the Qwen 3 prompt format. We recommend setting enable_thinking=True for all chats.

Here is an example inference script to get started:

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "ValiantLabs/Qwen3-8B-Esper3"

# 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 = "Write a Terraform configuration that uses the `aws_ami` data source to find the latest Amazon Linux 2 AMI. Then, provision an EC2 instance using this dynamically determined AMI ID."
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)

📄 License

This project is licensed under the Apache 2.0 license.

Esper 3 is created by Valiant Labs.

Check out our HuggingFace page to see all of our models!

We care about open source. For everyone to use.