The Hardware Cup final! 🏆
This is a AI post classified by AGK Intelligence as Open & local models (a model release), kept by the AI Radar because it carries real work, not commentary.
The Hardware Cup final! 🏆 RTX 5090 vs Mac Studio. Same AI, one war, one champion. Two semifinals, two winners, and now they fight for the cup. RTX 5090, 32 GB: Qwen3.8-27B, NVFP4 NInfer, MTP3 drafts Mac Studio M5 Ultra, 256 GB: Qwen3.8-27B, MLX 4-bit (4.695 bpw) TensorFold 1.2.3, DFlash2 drafts Same model, same game, one attempt, no help. Different quants and engines, so one match is a game, not a benchmark. NInfer and its model file by @neroued, with NVFP4 weights from @unslothai. TensorFold and its MLX checkpoint by @ashxhart. DFlash2 drafts by @inco_ai. Model by @Alibaba_Qwen.
Posted by Niklas (2.9k followers) 1 h ago · 9 likes · 314 views · view the original post on X. Kept by the AI Radar as Open & local models.
More AI work like this
- > GameFeel Bench has about 60,000 tasks about game balancing and understanding features — @teortaxesTex
- Got Arete 27B running blazing fast on my 10GB RTX 3080! — @PaulGug
- Another magic by DeepSeek v4.1 + Blender MCP + DeepSeek Harness running locally! — @ivanfioravanti
- ok this one's nerdy but I love it. Josef Gattermayer wrote AMD kernels for EXL3, the… — @MichaelGannotti
- 🥱 Tired of your #ai open-weight #LLM answering free speech questions like it's… — @michabbb
- You can now Run Uncensored DeepSeek-V4-Pro-Qwen3.5-9B locally on a 5 GB Laptop with zero… — @0x0SojalSec
- You can run locally Uncensored "DeepSeek-V4-Pro-Qwen3.5-9B" 5.6GB. — @0x0SojalSec
- Meet the Qwen3.8-27B Ultimate Heretic Uncensored Turbo Brilliance 50x AGI NEO GGUF. It's… — @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. 65.2k posts from 5k X accounts over the last 14 days, 10.6k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-08 00:39 UTC. Full method.