pinned
Paused
Agents
FLUX.2 [dev]
💻
Generate or edit images from text and optional photos
Lucas Silva PRO
YuLexuan30
AI & ML interests
None yet
Recent Activity
upvoted a changelog 16 days ago
MCP Server Enhancements reacted to SeaWolf-AI's post with 🚀 20 days ago
A small gift for anyone building or studying foundation models.
Most "open" models hand you the weights and stop there. With Aether-7B-5Attn we wanted to hand over the whole thing — so you can actually learn from it, reproduce it, and build on it: the data recipe, the training code, every hyperparameter, the complete logs, and the intermediate checkpoints. All Apache-2.0, reproducible byte-for-byte.
What you can do with it:
🔁 Rebuild it from scratch, or fork the recipe for your own model
🔬 Study a real heterogeneous-attention MoE — 49 layers place 5 attention mechanisms on a 7×7 Latin square, arranged as a clean, attributable ablation
📈 Trace training dynamics across the released checkpoints (110k / 115k / 162k)
It's a modest 6.59B model, and an honest one — the limitations (no KV-cache in this build, small scale) are written right in the card. We're not claiming it's special. If any piece of it saves you time or teaches you something, that's exactly what we hoped for. 🤗
📖 Full write-up →
[blog] · https://huggingface.co/blog/FINAL-Bench/opensource-llm
📦 5 Attention Base · https://huggingface.co/FINAL-Bench/Aether-7B-5Attn
🎯 5 Attention Instruct · https://huggingface.co/FINAL-Bench/Aether-7B-5Attn-it
🚀 5 Attention Live demo · https://huggingface.co/spaces/FINAL-Bench/Aether-Sovereign-AI
📦 7 Attention Base · https://huggingface.co/FINAL-Bench/Aether-7B-7Attn-base
📦 11 Attention Base · https://huggingface.co/FINAL-Bench/Aether-6B-11Attn-base
🧬 Collection · https://huggingface.co/collections/FINAL-Bench/aether-foundation-model
#opensource #LLM #MoE #reproducibility #Apache2Organizations
