Gemini 3.7 Flash Tops AA-AnalystAgent Benchmark
_philschmid · x · 2026-08-19
Gemini 3.7 Flash achieved #1 on the AA-AnalystAgent benchmark, which evaluates 80 real-world quantitative tasks. It scored 60% accuracy with the fastest speed (1.32s/task) and mid-range cost.
Related event: Gemini 3.7 Flash Tops AA-AnalystAgent Benchmark(2 posts)→
More from Models
- Mistral lags far behind frontier models in cyberdefense — emollick · 2026-08-19
- NVIDIA paper: Nemotron-H, a family of hybrid Mamba-Transformer models — cwolferesearch · 2026-08-19
- Ornith 1.5 released: 9B dense and 35B/397B MoE models — tarruda · 2026-08-19
- New Qwen Model Smaller and Easier to Run Than DeepSeek Flash — yacineMTB · 2026-08-19
- Z.ai's GLM-5.3 tops open-source rankings with lower price, delayed release — The Decoder · 2026-08-19
- GLM-5.3 tops EQ-Bench v4 with rare 9.1 analytical score — airesearch12 · 2026-08-19