River's open recipe beats GPT-6 Astra Pro and Claude Opus 5.5 on text-to-SQL at <1% cost
ibab · x · 2026-10-08
River AI launched River Recipes, an open-source initiative turning research into vetted, reproducible training recipes. Its first recipe (RL + LoRA on Kimi K2.6, rebuilt from ReViSQL) pushes four open models to human-level 92.96% on Arcwise-Plat text-to-SQL, beating GPT-6 Astra Pro by 4 points at <2% of the cost, with GLM-5.3 Flash ahead by 4.6 points at just 0.5% cost.
Related event: River Open-Sources Training Recipes to Match GPT-6 at Under 1% of the Cost(2 posts)→
More from Research
- Perplexity releases open multimodal late-interaction embedding family with SOTA results — antoine_chaffin · 2026-10-08
- User challenges frontier AI labs to drop 700+ cancer-cure papers as the true AGI milestone — BLUECOW009 · 2026-10-08
- Paper: Vibe Coding Kills Open Source as AI-Recommended Repos Lose Stars — soumitrashukla9 · 2026-10-08
- Scaling AI-guided experiments beats scaling biological data, researcher argues at ICML — anshulkundaje · 2026-10-08
- MA-BC: Provably Efficient Multi-Objective Imitation Learning from Heterogeneous Experts — Yossarian_1234 · 2026-10-08
- Hide Model Names From Agents: Labels Cost 55% More Tokens and Drop Success to 81% — alex_verem · 2026-10-08