Nemotron 3.5 Super VL Hits River API, Beats Opus 5.5 on SQL After Fine-tuning
ibab · x · 2026-10-15
- River partnered with NVIDIA to make Nemotron 3.5 Super VL available on the River API from day zero, supporting inference, fine-tuning, and RL.
- Out of the box, Super VL outperforms Kimi K2.6 on River's text-to-SQL eval despite being a fraction of its size.
- After training with River's text-to-SQL recipe, it beat every proprietary model evaluated, including Astra and Opus 5.5.
- Takeaway: small models, when fine-tuned for a specialized task, can reach frontier-level performance while remaining deployable on your own hardware.
More from Models
- Reasonable post-trains Nemotron 3.5 on 3,000 Verus proofs, lifting pass@1 ~30 points to 73.3% — fhuszar · 2026-10-16
- NVIDIA's Nemotron 3.5 Super VL hits CoreWeave: 12B-active multimodal MoE at $0.25/M input tokens — wandb · 2026-10-16
- Opus 5 might be a very deep model or use looping, commentator speculates — scaling01 · 2026-10-16
- GPT-6.1 Sol audit finds 6% of arXiv papers had false main theorems, down to 1.5% recently — nitarshan · 2026-10-16
- Nvidia-backed post-training pushes Nemotron 3.5 VL past frontier on dermatology diagnosis — LoganGrasby · 2026-10-16
- Gemini 4 Argon scores 97.5% on ARC-AGI-3, second lab above 95% — patience_cave · 2026-10-16