FINAL-Bench / Aether-7B-5Attn-it

huggingface.co
Total runs: 168
24-hour runs: 0
7-day runs: -28
30-day runs: -23
Model's Last Updated: July 23 2026
text-generation

Introduction of Aether-7B-5Attn-it

Model Details of Aether-7B-5Attn-it

Aether-7B-5Attn-it

Aether-7B-5Attn

📚 Part of the Aether Foundation Model collection — base, instruction-tuned, and checkpoints in one place. Instruction-tuned (SFT) version of the fully-open Aether-7B-5Attn base model. Post-trained for multiple-choice / benchmark-style answering. 6.59B MoE (~2.98B active), 49 layers on a 7×7 Latin square. Apache-2.0.

Learning rate chosen by held-out accuracy, not loss

Full-parameter SFT was run at three learning rates (2e-6 / 6e-6 / 2e-5). The final checkpoint was selected by held-out benchmark accuracy, not training loss — for small models a high LR lowers training loss while degrading real capability, so loss is the wrong selector.

LR K-AI 4 avg Note
2e-6 29.7% Under-fit (weak on some subjects)
6e-6 (selected) 34.9% Best & balanced, no degradation
2e-5 32.3% One subject collapsed (format-degradation sign)
Held-out results (selected checkpoint, 6e-6)
  • GPQA-Diamond (198): 25.3%
  • K-AI 4 average (195): 34.9%
    • musr_ko 26.5% · com2_main_ko 50.0% · click 32.0% · kommlu_pro 30.4%
  • vs base (before SFT): base 26.7% → 34.9% (+8.2pp) — same held-out set, same harness.

All evaluations use held-out sets not seen in training. SFT was performed on MMLU-auxiliary (multiple-choice format), which does not overlap with the evaluation subjects (GPQA / K-AI).

Pretraining data mix (inherited from base)
Domain Share
Math (finemath + open-web-math) 37.8%
Korean (webtext + synth) 21.6%
English web & synthetic (fineweb-edu + cosmopedia) 21.6%
Code (opc) 13.5%
phase15 pre-blend 5.4%

This is the base pretraining mix. This model's post-training (SFT) data is MMLU-auxiliary.

Architecture (inherited from base)

Identical to Aether-7B-5Attn base: 49 layers placed on a 7×7 Latin square with heterogeneous attention (7 labels / 5 distinct mechanisms) and a 25-expert MoE (top-7 + 1 shared). Full structural detail and the diagram are in the base card, §3.2: FINAL-Bench/Aether-7B-5Attn .

Open-source fully-open LLMs — 6-country comparison

Open-source LLM comparison

Relative to the base Aether. Among six sovereign fully-open models, VIDRAFT is the only single AI startup, and Aether has the most attention types (5) in a Latin-square layout.

Usage

Uses the custom aether_pkg/ architecture ( trust_remote_code=True ). Follow the loading instructions in the base model card .

⚠️ Use batch_size=1. Some attention branches (NSA family) do not consume a padding mask, so batching padded sequences can silently corrupt results. Run inference one sequence at a time.

Contact

VIDRAFT (주식회사 비드래프트) · [email protected] · License: Apache-2.0

Runs of FINAL-Bench Aether-7B-5Attn-it on huggingface.co

168
Total runs
0
24-hour runs
-16
3-day runs
-28
7-day runs
-23
30-day runs

More Information About Aether-7B-5Attn-it huggingface.co Model

More Aether-7B-5Attn-it license Visit here:

https://choosealicense.com/licenses/apache-2.0

Aether-7B-5Attn-it huggingface.co

Aether-7B-5Attn-it huggingface.co is an AI model on huggingface.co that provides Aether-7B-5Attn-it's model effect (), which can be used instantly with this FINAL-Bench Aether-7B-5Attn-it model. huggingface.co supports a free trial of the Aether-7B-5Attn-it model, and also provides paid use of the Aether-7B-5Attn-it. Support call Aether-7B-5Attn-it model through api, including Node.js, Python, http.

Aether-7B-5Attn-it huggingface.co Url

https://huggingface.co/FINAL-Bench/Aether-7B-5Attn-it

FINAL-Bench Aether-7B-5Attn-it online free

Aether-7B-5Attn-it huggingface.co is an online trial and call api platform, which integrates Aether-7B-5Attn-it's modeling effects, including api services, and provides a free online trial of Aether-7B-5Attn-it, you can try Aether-7B-5Attn-it online for free by clicking the link below.

FINAL-Bench Aether-7B-5Attn-it online free url in huggingface.co:

https://huggingface.co/FINAL-Bench/Aether-7B-5Attn-it

Aether-7B-5Attn-it install

Aether-7B-5Attn-it is an open source model from GitHub that offers a free installation service, and any user can find Aether-7B-5Attn-it on GitHub to install. At the same time, huggingface.co provides the effect of Aether-7B-5Attn-it install, users can directly use Aether-7B-5Attn-it installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

Aether-7B-5Attn-it install url in huggingface.co:

https://huggingface.co/FINAL-Bench/Aether-7B-5Attn-it

Url of Aether-7B-5Attn-it

Aether-7B-5Attn-it huggingface.co Url

Provider of Aether-7B-5Attn-it huggingface.co

FINAL-Bench
ORGANIZATIONS

Other API from FINAL-Bench

huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:October 19 2026