Google DeepMindGemini
Gemini 2.0 Flash
Google DeepMind · Gemini. Released Dec 11, 2024.
Gemini 2.0 Flash is a model from Google DeepMind in the Gemini family, released Dec 11, 2024. evals.report tracks 22 reported Gemini 2.0 Flash benchmark scores across GPQA Diamond, FrontierMath, AIME (OTIS Mock), Epoch Capabilities Index, Aider Polyglot, MMLU-Pro, GAIA: A Benchmark for General AI Assistants, MathVista, and 14 more — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.
22 results
Benchmark results 22
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| GPQA Diamond | Reasoning | 64.1% | accuracy | Official | Dec 11, 2024 | Details |
| FrontierMath | Reasoning | 1.72% | accuracy | Official | Dec 11, 2024 | Details |
| AIME (OTIS Mock) | Reasoning | 31.1% | accuracy | Official | Dec 11, 2024 | Details |
| Epoch Capabilities Index | Reasoning | 135.9 | Index | Official | Dec 11, 2024 | Details |
| Aider Polyglot | Coding | 22.2% | % correct | Official | Dec 11, 2024 | Details |
| MMLU-Pro | Reasoning | 77.9% | accuracy | Verified | Dec 11, 2024 | Details |
| GAIA: A Benchmark for General AI Assistants | Agents | 32.73% | accuracy | Unverified | Dec 11, 2024 | Details |
| MathVista | Multimodal | 73.1% | accuracy | Unverified | Dec 11, 2024 | Details |
| GDPval | Agents | 566 | Elo | Official | Dec 11, 2024 | Details |
| LiveCodeBench | Coding | 33.4% | Pass@1 | Unverified | Dec 11, 2024 | Details |
| SciCode | Coding | 33.3% | accuracy | Unverified | Dec 11, 2024 | Details |
| MMMU (Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark) | Multimodal | 70.7% | accuracy | Unverified | Dec 11, 2024 | Details |
| MultiChallenge | Reasoning | 36.35% | accuracy | Verified | Dec 11, 2024 | Details |
| LongBench v2 | Reasoning | 51.1% | accuracy | Official | Dec 11, 2024 | Details |
| EQ-Bench Creative Writing v3 | Chat preference | 1239 | Elo | Verified | Dec 11, 2024 | Details |
| AILuminate AI Safety Benchmark | Other | Good | Safety grade | Verified | Dec 11, 2024 | Details |
| MASK (Model Alignment between Statements and Knowledge) | Other | 49.07 | Honesty score | Verified | Dec 11, 2024 | Details |
| FACTS Grounding | Reasoning | 83.6% | Grounding accuracy | Verified | Dec 11, 2024 | Details |
| BigCodeBench | Coding | 33.8% | calibrated Pass@1 | Verified | Dec 11, 2024 | Details |
| SuperGPQA | Reasoning | 47.73% | accuracy | Verified | Dec 11, 2024 | Details |
| EnigmaEval | Reasoning | 0.63% | accuracy | Verified | Dec 11, 2024 | Details |
| Online-Mind2Web | Agents | 29.00% | Task success rate | Verified | Dec 11, 2024 | Details |