VIBE (MiniMax) - Backend: leaderboard

The Backend subset of MiniMax's VIBE: the model, run by MiniMax, builds server-side systems from a written requirement, and the pipeline generates and runs test scripts against them; the score averages three runs. MiniMax does not say how many tasks each model ran; the released set has 40 Backend prompts.

Metric: Score (0-100). Source: github.com. Saturation forecast: Estimated already saturated. 5 models tracked. Updated 17 Nov 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 4.598#161
2Claude Sonnet 4.590.8#258
3MiniMax-M2.186.7#573
4Gemini 3 Pro78.7#145
5MiniMax-M267.8#687

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=vibe-minimax-backend · How It Works · Data refreshed daily, snapshot 2026-11-17.