[whatmodel.sh](../../README.md) / Gemini 2.0 Flash Experimental # Gemini 2.0 Flash Experimental Google · released 2024-12 · proprietary **score 56.8** ±1.3 · comparison only: no current run option verified · from 13 benchmark results across 4 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at ● marks Gemini 2.0 Flash Experimental among every model on that task or test, worst to best. 
task    vs the others         rank  counts  tests
coding  ▁▁▁▃▄▆●▇▅▄▃▁   #245 of 539     68%      3
coding · #245 of 539 · 68% of its score Beats [Qwen3.5 4B](qwen3-5-4b.md), [Falcon-H1R 7B](falcon-h1r-7b.md), [Qwen3-VL 32B Instruct](qwen3-vl-32b-instruct.md) here, though they rank higher overall. Trails [QwQ-32B-Preview](qwq-32b-preview.md), [Ministral 3 3B Reasoning 2512](ministral-3-3b-reasoning.md), [DeepSeek-Coder-V2 Instruct (0724)](deepseek-coder-v2-0724.md) here, though they rank lower overall. 
test                                result  vs the others in its run  difficulty            counts  why
Aider polyglot (225 exercises, tw…   22.2%  ▂·▂▃●▂▃█▆▄▂▂ beats 13/49  very hard · avg 7.6%     36%  sharp
Aider code editing (133 Exercism …   69.9%  ▁·▁▂▄▁▆▄█●▂▂ beats 47/62  medium · avg 51%         31%  sharp · at its level
LiveCodeBench official release_v5…   37.9%  ●···█····█·█ beats 0/3    medium · avg 51%        1.7%  at its level · noisy
other tests (10) 
test                                result  vs the others in its run  difficulty           counts  why
AHELM Speech Robust Bench (LibriS…   17.1%  ▂·█··▆●·▂▂·▄ beats 9/13   avg 34%                 13%  typical
AHELM CoRe-Bench conversational a…   75.4%  ▂·▄··▂▂·▄▂●▄ beats 9/13   medium · avg 52%       6.1%  at its level
AHELM LibriSpeech transcription b…    4.4%  ▃···▃·█▃▅▃▅● beats 11/13  avg 11%                3.2%  typical
AHELM VoxCeleb2 same-speaker dete…   53.0%  ▂·▂·▂●▂▄▄··▄ beats 6/13   very hard · avg 54%    3.2%  sharp · noisy
AHELM MUStARD audio sarcasm detec…   60.4%  ▂·······▆█●▆ beats 10/13  very hard · avg 57%    2.6%  blunt
AHELM AIR-Bench Foundation music …   69.8%  ▃█▃▅▃·▃·▃●▃▃ beats 11/13  medium · avg 57%       1.2%  blunt · at its level
AHELM MELD audio emotion classifi…   44.3%  ▃··▃▃▃·▃●▅█▅ beats 6/13   medium · avg 39%       1.0%  blunt · at its level
AHELM FLEURS 7-language transcrip…   64.6%  ▄▂···▂●▂█▂·▂ beats 6/13   avg 83%                0.6%  blunt · noisy
AHELM Multilingual LibriSpeech 7-…    6.0%  ▂▂▂··▂▂·▂▂█● beats 11/13  avg 14%                0.1%  blunt · noisy
AHELM VocalSound human-sound reco…   71.8%  ██▄▄·●·█·█▄▄ beats 6/13   easy · avg 67%         0.0%  blunt · at its level · noisy
## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its *counts* column above; here it is added up by who ran the tests. [Method](../method.md) 
source                                    results  share of the score
Aider polyglot benchmark runs                   1   36% ▇▇▇▇
AHELM: A Holistic Evaluation of Audio-L…       10   31% ▇▇▇▇
Aider code editing benchmark runs               1   31% ▇▇▇▇
LiveCodeBench official website results          1    2% ▇
prior: pull toward the middle                        1% ▇
weighted average                                   = 56.8
Licences: Aider polyglot benchmark runs ([Apache-2.0](../../licenses/aider-polyglot-runs-Apache-2.0.txt)) · AHELM: A Holistic Evaluation of Audio-Language Models ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Aider code editing benchmark runs ([Apache-2.0](../../licenses/aider-polyglot-runs-Apache-2.0.txt)) · LiveCodeBench official website results ([CC-BY-SA-4.0](../../licenses/lcb-sharealike-NOTICE.txt)) ## How to run it No verified API or local download yet. [interactive version](https://whatmodel.sh/model/gemini-2-0-flash-exp) · [JSON](https://whatmodel.sh/api?format=json&all=1)