Instructions to use FINAL-Bench/Aether-7B-5Attn with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FINAL-Bench/Aether-7B-5Attn with Transformers:

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

pipe = pipeline("text-generation", model="FINAL-Bench/Aether-7B-5Attn", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("FINAL-Bench/Aether-7B-5Attn", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use FINAL-Bench/Aether-7B-5Attn with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "FINAL-Bench/Aether-7B-5Attn"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "FINAL-Bench/Aether-7B-5Attn",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/FINAL-Bench/Aether-7B-5Attn

SGLang

How to use FINAL-Bench/Aether-7B-5Attn 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 "FINAL-Bench/Aether-7B-5Attn" \
    --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": "FINAL-Bench/Aether-7B-5Attn",
		"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 "FINAL-Bench/Aether-7B-5Attn" \
        --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": "FINAL-Bench/Aether-7B-5Attn",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use FINAL-Bench/Aether-7B-5Attn with Docker Model Runner:
```
docker model run hf.co/FINAL-Bench/Aether-7B-5Attn
```

SeaWolf-AI commited on Jul 19

Commit

7636fdf

verified ·

1 Parent(s): 0242953

card: add hero image + comparison table image

Browse files

Files changed (1) hide show

README_ko.md +10 -0

README_ko.md CHANGED Viewed

@@ -25,6 +25,9 @@ datasets:
 # Aether-7B-5Attn — Base
 ## 🇰🇷 학습 데이터와 학습 코드까지 전부 공개하는 한국 최초의 파운데이션 모델
 **49층에 어텐션을 7×7 라틴 방진으로 배치한 6.59B MoE 언어모델.**
@@ -66,6 +69,13 @@ NVIDIA **B200 16장**(2-node FSDP)으로 **약 46일**, 총 **144.2B 토큰**을
 ---
 ## 2. 설계 — 왜 모든 층이 같은 어텐션이어야 하는가
 ### 2.1 하나의 어텐션이 모든 층을 담당해야 할 이유는 없습니다

 # Aether-7B-5Attn — Base
+![Aether-7B-5Attn](aether_hero.png)
 ## 🇰🇷 학습 데이터와 학습 코드까지 전부 공개하는 한국 최초의 파운데이션 모델
 **49층에 어텐션을 7×7 라틴 방진으로 배치한 6.59B MoE 언어모델.**
 ---
+### 오픈소스 완전공개 LLM 비교 (6개국)
+![Open-source LLM comparison](opensource_llm_comparison.png)
+*소버린 완전공개 LLM 6개국 대비. 개방 항목은 모두 대등하며, 비드래프트의 차별점은 유일한 한국 AI 스타트업이라는 점과 5종(최다) 어텐션·라틴방진 구조입니다.*
 ## 2. 설계 — 왜 모든 층이 같은 어텐션이어야 하는가
 ### 2.1 하나의 어텐션이 모든 층을 담당해야 할 이유는 없습니다