Releasing Delta League: edit LLM weights for benchmarking and glory! Join at…
This is a AI post classified by AGK Intelligence as AI research (a launch), kept by the AI Radar because it carries real work, not commentary.
Releasing Delta League: edit LLM weights for benchmarking and glory! Join at deltaleague(.)ai Earlier this week, the idea of editing weights directly with an LLM got a lot of attention — but it was shunted by limitations in the demonstrated approach. Delta League is an open competition for everyone to submit deltas to model weights and get them scored on a hidden test set. Auto-research and hillclimb to make smaller models more performant. The broader vision for Delta League is creating a platform for open AI research. Beyond just editing weights, this approach offers verifiable evaluation
Posted by Illia (root.near) (🇺🇦, ⋈) (121.7k followers) 1 days ago · 300 likes · 39.6k views · view the original post on X. Kept by the AI Radar as AI research.
More AI work like this
- No need for me to manually scan for new model releases anymore, since @paperswithcodee… — @NielsRogge
- We’re introducing ToyBlocks, an open simulated tabletop for controlled experiments in… — @allen_ai
- HotBench - Expert-authored tasks in medicinal chemistry, microbiology and dermatology to… — @SynBio1
- I wonder if you realize how interesting this is. — @SciTechera
- 🔥我靠！终于有大公司把AI模型后训练的“魔法”拆成可复现配方了！ — @NFT_Chen
- BREAKING: the best of 8 frontier models scored 37.5/100 on Science's Last Exam. On true… — @guilleflorvs
- Stanford dropped a 2-hour course on building LLMs from scratch — @RoundtableSpace
- Meet DINOv3 ViT-H/16+: a powerful image feature extractor from the DINOv3 family. With… — @HuggingModels
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.6k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 16:05 UTC. Full method.