MiniMax M3.1 Flash Hits 73.3 on BuildingBench, Reshaping the Cost-Quality Frontier
Lianhuiq · x · 2026-10-20
Third-party evals show MiniMax M3.1 Flash scoring 73.3 overall on the 3D BuildingBench (rank #12), beating SWE-2's 67.3 to claim a spot on the cost-quality Pareto frontier and nearing DeepSeek V4.1 Flash's 72.9—currently free via OpenRouter. Its best reconstruction hits 88.8 on Melbourne's Australian Centre for Contemporary Art.
More from Models
- Hazy Research's open-source NCM generates text 31x faster via coarse-to-fine concept modeling — simran_s_arora · 2026-10-20
- Blogger claims DeepSeek's unreleased model nearly matches Opus at design, shares one-prompt demo — sven_ai · 2026-10-20
- Fingerprint test flags DeepSeek ds-4.1-flash as likely GPT-6-astra — sven_ai · 2026-10-20
- Intelligence Saturation: 10-Point Model Edge Now Buys Just 20% More Developer Traffic — khademinori · 2026-10-20
- User: Codex burns half of the day's usage limits within an hour after banked reset — ssh4net · 2026-10-20
- Language Modeling Is Agents in the Limit: voooooogel's Intuition Clash Sparks Debate — voooooogel · 2026-10-20