Today for AI
HOT RADAR
llmHEAT 26.7°

AI CURATED HOT STORY · 11/6/2026

BOSS Zhipin's Nanbeige Team Releases Open-Weights Nanbeige4.3-33B, Matching Qwen3.8 27B on Intelligence Index

1 reports archived1 independent sourcesupdated 11/6/2026, 01:49:19
Synthesis & Latest Updates
1 Sources Cross-Validated

Nanbeige, the foundation model team at Chinese recruitment platform BOSS Zhipin, has released Nanbeige4.3-33B under Apache 2.0 — a dense 33B reasoning model scoring 20 on the Artificial Analysis Intelligence Index v5, level with Qwen3.8 27B (xhigh). It offers a 524K-token context window, text-only I/O, and is small enough to run locally on a laptop, with terminal use and physics reasoning as relative strengths but weaker knowledge benchmarks.

LATEST/Nanbeige, the foundation model team at Chinese recruitment platform BOSS Zhipin, has released Nanbeige4.3-33B under Apache 2.0 — a dense 33B reasoning model scoring 20 on the Artificial Analysis Intelligence Index v5, level with Qwen3.8 27B (xhigh). It offers a 524K-token context window, text-only I/O, and is small enough to run locally on a laptop, with terminal use and physics reasoning as relative strengths but weaker knowledge benchmarks.

Hourly heat curve

2 fully observed hours
Fewer than 3 fully observed hours; trend not plotted yet.

Coverage timeline

Total 1 reports · Latest first
  1. Artificial Analysis (@ArtificialAnlys)T1·72 pts

    BOSS Zhipin's Nanbeige Team Releases Open-Weights Nanbeige4.3-33B, Matching Qwen3.8 27B on Intelligence Index

    • Nanbeige4.3-33B ties Qwen3.8 27B (xhigh) at 20 on AA Intelligence Index v5, far ahead of same-scale Muse Glimmer at 6
    • Scores 30% on Terminal-Bench 4.0 and 16% on CritPt v2 Hard Subset, versus 6% and 4% for Qwen3.8 27B
    • Trails on knowledge-heavy benchmarks: 1162 vs 1391 Elo on AA-Briefcase v1.1, 7% vs 15% on HLE-Diamond, 11% vs 15% on FrontierCode 2.0
    • Apache 2.0 open weights, 524K context, text-only, with no confirmed public API