Nanbeige4.3-33B open-sourced: claims SOTA agentic performance among ~30B open models
victormustar · x · 2026-10-25
Nanbeige released Nanbeige4.3-33B, an open model built with a 'strong teacher first' strategy to push small-model agent performance.
- Claims SOTA agentic coding and cowork benchmark results among open models at 30B parameters
- Works inside mainstream agent harnesses on real-world tasks with cost-efficient inference
- Will serve as the teacher for upcoming smaller models, starting with Nanbeige4.3-3B
Weights are up on Hugging Face.
Related event: Nanbeige4.3-33B Open-Sourced, Claimed to Beat Qwen3-32B on Benchmarks(4 posts)→
More from Models
- Meta Muse avatar sharing surfaces: four avatar states, social sharing, no release date — testingcatalog · 2026-10-25
- Meta Muse to add avatar sharing, hinting at an agent avatar store — testingcatalog · 2026-10-25
- Skeptic doubts rumored OpenAI model, citing inability to serve Astra reliably — CtrlAltDwayne · 2026-10-25
- AsciiBench: blind voting arena tests which AI draws better ASCII art across 13 models — JasonBotterill · 2026-10-25
- Satirical ARC-AGI-3 leaderboard shows GPT-6 near 90% score at $10K+ per task — haider1 · 2026-10-25
- Rumor: Anthropic to drop Opus 5.5 two weeks after calling for frontier pacing — StewartalsopIII · 2026-10-25