9B Coding Model Trained on Bittensor Beats Qwen Flagship on SWE-bench

Reliquary AI released Teutonic-I-Reliquary, a 9B model pretrained on Teutonic (3) and post-trained in Bittensor's competitive environment, scoring 43.6% on SWE-bench and outperforming Qwen's flagship 9B model.

2026-11-05 ~ 2026-11-06 · 2 related posts

• 9B Model Trained on Bittensor Scores 43.6% on SWE-bench, Beating Qwen3.5-9B — const_reborn · 2026-11-05
• Bittensor-trained 9B model beats Qwen's flagship 9B at coding using competitively sourced anonymous compute — const_reborn · 2026-11-06