ModelsAug 22, 2026 · 21:485 sources

Ox Alpha Hits 80% on Full DeepSWE Benchmark, First Run of All 113 Tasks

From huggingnews.com

Image: huggingnews.com
AI;DR

The stealth AI model Ox Alpha achieved an approximately 80% pass rate on a full run of 113 tasks in the DeepSWE coding benchmark. The model features a 1M token context window and supports multimodal input across text, image, and video. It is currently available for free on the OpenRouter, Venice, and Nous Research platforms.

Read the sourceSign in to suggest

Sources

+ 3 more
  1. SUPPORT@rohanpaul_ai“One prompt, 64,745 output tokens, and no external assets: Ox Alpha generated the code for an entire Three.js dreamcore 3D scene/world in a single shot”x.com
  2. SUPPORT@teortaxestex“The 100T/day promise (unsubstantiated so far afaik) is lowkey genius”x.com
  3. SUPPORT@teortaxestex“it does not take a lot of intelligence to perform SWE work at a superhuman level”x.com

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today