AI Radar
Support
LiveUpdated 2026-11-06 16:05 UTC

Final push for the TensorFold http://cuda.fast and http://mlx.fast challenges with…

Let's keep going. 🚀 Final push for the TensorFold http://cuda.fast and http://mlx.fast challenges with Darkbloom,…

This is a AI post classified by AGK Intelligence as AI infra & evals (a model release), kept by the AI Radar because it carries real work, not commentary.

Let's keep going. 🚀 Final push for the TensorFold http://cuda.fast and http://mlx.fast challenges with Darkbloom, Yukon and Eigen Labs. Entries close at 05:00 UTC on Saturday, 7 November. Current Qwen 3.8 27B records: http://cuda.fast: +175.9% on DGX Spark. http://mlx.fast: +81% on Apple M5 Max. Those are weighted gains against each track's reference, 75% decode and 25% prefill. If you've got an optimisation to try, get a run submitted. Let's see how much further we can push TensorFold before it closes. Thank you to everyone taking part. CUDA challenge · MLX/Metal challenge @0xkydo @Gum

Posted by Ash Hart (11.1k followers) 1 days ago · 42 likes · 2.8k views · view the original post on X. Kept by the AI Radar as AI infra & evals. Tools mentioned: cuda.fast, mlx.fast.

More AI work like this

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.