Nexusflow / athene family / text-generation

Athene-70B

Nexusflow/Athene-70B

other SAFETENSORS 70.6B params Provisional scores Updated 15 Nov 2024

Benchmarks

Where it scores.

Methodology →

General Provisional

—

Coding Provisional

—

Math Provisional

—

Reasoning Provisional

—

Normalized 0–100 across indexed models.

Runnable variants

Pick a size.

30 indexed

Precision	Format	Size	Params	File
FP16 / smallest	SAFETENSORS	2.1 GB	—	model-00030-of-00030.safetensors
FP16	SAFETENSORS	4.6 GB	—	model-00001-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00005-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00010-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00015-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00020-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00025-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00002-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00006-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00007-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00011-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00012-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00016-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00017-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00021-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00022-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00026-of-00030.safetensors
FP16	SAFETENSORS	4.7 GB	—	model-00027-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00004-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00009-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00014-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00019-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00024-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00029-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00003-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00008-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00013-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00018-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00023-of-00030.safetensors
FP16	SAFETENSORS	5.0 GB	—	model-00028-of-00030.safetensors

Sizes are file sizes, not peak runtime memory.

From the Hugging Face model card

About this model.

Llama3-Athene-70B

We introduce Llama3-Athene-70B, an open-weights LLM trained through RLHF based off Llama-3-70B-Instruct. Athene-70B achieves a high score on Arena-Hard-Auto, a proxy benchmark for Chatbot Arena.

Developed by: The Nexusflow Team (Evan Frick*, Peter Jin*, Tianle Li*, Karthik Ganesan, Jian Zhang, Jiantao Jiao and Banghua Zhu).
Model type: Chat Model
Finetuned from model: Llama-3-70B-Instruct.
License: Nexusflow Research License
Blog: https://nexusflow.ai/blogs/athene

Model	Arena-Hard
Claude-3.5-Sonnet (Proprietary)	79.3%
GPT-4o (Proprietary)	79.2%
Athene-70B (Open)	77.8%
Gemini-Pro-1.5 (Proprietary)	72.0%
Gemma-2-27B (Open)	57.0%
Llama-3-70B (Open)	46.6%

Usage

Athene-70B uses the same chat template as Llama-3-70B-Instruct. Below is an example simple usage using the Transformers library.

import transformers
import torch

model_id = "Nexusflow/Athene-70B"

pipeline = transformers.pipeline(
    "text-generation",
    model=model_id,
    model_kwargs={"torch_dtype": torch.bfloat16},
    device_map="auto",
)

messages = [
    {"role": "system", "content": "You are an Athene Noctura, you can only speak with owl sounds. Whoooo whooo."},
    {"role": "user", "content": "Whooo are you?"},
]

terminators = [
    pipeline.tokenizer.eos_token_id,
    pipeline.tokenizer.convert_tokens_to_ids("<|end_of_text|>")
]

outputs = pipeline(
    messages,
    max_new_tokens=256,
    eos_token_id=terminators,
    do_sample=True,
    temperature=0.6,
    top_p=0.9,
)
print(outputs[0]["generated_text"][-1])

Acknowledgment

We would like to thank the LMSYS Organization for their support of testing the model. We would like to thank Meta AI and the open source community for their efforts in providing the datasets and base models.

Citation

@misc{Athene2024,
    title = {Athene-70B: Redefining the Boundaries of Post-Training for Open Models},
    url = {https://nexusflow.ai/blogs/athene},
    author = {Frick, Evan and Jin, Peter and Li, Tianle and Ganesan, Karthik and Zhang, Jian and Jiao, Jiantao and Zhu, Banghua},    
    month = {July},
    year = {2024}
}