How to use from
Docker Model Runner
docker model run hf.co/EvaryxTechnologies/Evaryx-Core
Quick Links

Evaryx-Core (8B)

Evaryx-Core is an 8-billion-parameter conversational and reasoning language model developed and fine-tuned by EVARYX TECHNOLOGIES. Built on the Qwen2 architecture, it is trained to deliver natural, human-like responses alongside structured, deep-thinking problem-solving.

Key Capabilities

  • Natural Communication: Responds in a clear, fluid, and engaging tone without sounding robotic.
  • Deep Thinking: Breaks down complex questions, logic puzzles, and multi-step problems thoughtfully.
  • General Problem Solving: Assists with coding, brainstorming, explanations, and everyday decision-making.

Model Details

  • Developed by: EVARYX TECHNOLOGIES
  • Model Type: Causal Language Model (Fine-tuned)
  • Architecture: Qwen2 (8B Parameters)
  • Precision: F16 (Safetensors)

Quickstart (Python)

Option 1: Standard Loading (16GB+ VRAM)

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "EvaryxTechnologies/Evaryx-Core"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto"
)

messages = [
    {"role": "user", "content": "Help me think through a complex problem step by step."}
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
print(response[0])

Option 2: Low-Memory 4-Bit Loading (8GB VRAM)

Requires bitsandbytes (pip install -U bitsandbytes accelerate transformers).

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig

model_id = "EvaryxTechnologies/Evaryx-Core"

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_compute_dtype="float16"
)

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    quantization_config=bnb_config,
    device_map="auto"
)

messages = [
    {"role": "user", "content": "Help me think through a complex problem step by step."}
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
print(response[0])

Intended Use & Limitations

Evaryx-Core is built for research, assistant applications, and general problem-solving. As with all language models, outputs should be verified for critical factual, medical, legal, or financial decisions.

Downloads last month
245
Safetensors
Model size
8B params
Tensor type
F16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for EvaryxTechnologies/Evaryx-Core

Quantizations
2 models