whatmodel.sh / Gemini 2.0 Flash Experimental
Gemini 2.0 Flash Experimental
↗ share
+ contribute
Gemini 2.0 Flash Experimental is a closed-weights language model from Google, released December 2024. Licence: proprietary.
Catalogue identity notes
Comparison-only: retired or no longer listed API model, kept so its independent runs can be scored. Gemini API changelog, December 11, 2024: 'Released Gemini 2.0 Flash Experimental for public preview' (model id gemini-2.0-flash-exp); the changelog's shutdown list (December 9, 2025) includes gemini-2.0-flash-exp. Distinct from the GA gemini-2.0-flash-001 (catalogued). Runs below are dated December 2024.
$ how it compares
The LLMs ranked either side of Gemini 2.0 Flash Experimental, out of 828 with a score, then more from Google.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 289 | Gemini 1.5 Pro (002) | 56±1 | comparison only | ||
| 290 |  Athene-V2-Chat 72B | 56±2 | local | ||
| 291 | Gemini 2.0 Flash Experimental | 56±1 | comparison only | ||
| 292 |  Sky-T1 32B Preview | 56±5 | local | ||
| 293 |  Qwen Plus (2025-01-25) | 56±1 | api $0.60/M | ||
| more from Google | |||||
| 9 | Gemini 4 Argon (limited preview) | 82±1 | limited preview | ||
| 23 | Gemini 3.8 Flash | 77±1 | api $1.50/M | ||
| 32 | Gemini 3.7 Flash | 76±1 | api $1.50/M | ||
$ what it's good at
coding▁▁▁▃▅▇█▇▅▄▃▁ #238 of 524 · 66% of its score · 3 tests
Beats Qwen3.5 4B, Qwen3-VL 32B Instruct, Falcon-H1R 7B here, though they rank higher overall. Trails QwQ-32B-Preview, Ministral 3 3B Reasoning 2512, DeepSeek-Coder-V2 Instruct (0724) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Aider polyglot (225 exercises, two attempts) | 22.2% ✎ | ▂·▂▃▅▂▃█▆▄▂▂ beats 13/49 | 36% |
| Aider code editing (133 Exercism Python exercises, two attempts) | 69.9% ✎ | ▁·▁▂▄▁▆▄█▆▂▂ beats 47/62 | 28% |
| LiveCodeBench official release_v5 (Nov 1, 2023–Feb 1, 2025) | 37.9% ✎ | █···█····█·█ beats 0/3 | 1.7% |
other tests10 not in a task · 33% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| AHELM Speech Robust Bench (LibriSpeech-clean with noise) transcription WER | 17.1% ✎ | ▂·█··▆▄·▂▂·▄ beats 9/13 | 16% |
| AHELM CoRe-Bench conversational audio reasoning (pseudo-exact match) | 75.4% ✎ | ▂·▄··▂▂·▄▂█▄ beats 9/13 | 5.2% |
| AHELM VoxCeleb2 same-speaker detection (exact match) | 53.0% ✎ | ▂·▂·▂█▂▄▄··▄ beats 6/13 | 3.3% |
| AHELM LibriSpeech transcription by audio LLMs (WER) | 4.4% ✎ | ▃···▃·█▃▅▃▅█ beats 11/13 | 3.3% |
| AHELM MUStARD audio sarcasm detection (exact match) | 60.4% ✎ | ▂·······▆█▆▆ beats 10/13 | 2.5% |
| AHELM AIR-Bench Foundation music subsets (exact match) | 69.8% ✎ | ▃█▃▅▃·▃·▃▅▃▃ beats 11/13 | 1.4% |
| AHELM MELD audio emotion classification (pseudo-exact match) | 44.3% ✎ | ▃··▃▃▃·▃▅▅█▅ beats 6/13 | 0.9% |
| AHELM FLEURS 7-language transcription by audio LLMs (WER) | 64.6% ✎ | ▄▂···▂▆▂█▂·▂ beats 6/13 | 0.5% |
2 more tests
| test | result | vs the others in its run | counts |
|---|---|---|---|
| AHELM Multilingual LibriSpeech 7-language transcription by audio LLMs (WER) | 6.0% ✎ | ▂▂▂··▂▂·▂▂█▆ beats 11/13 | 0.1% |
| AHELM VocalSound human-sound recognition (prefix exact match) | 71.8% ✎ | ██▄▄·█·█·█▄▄ beats 6/13 | 0.0% |
$ where to run it
No verified API or download recorded yet · add one
$ how the score was worked out · Aider polyglot benchmark runs 36%, AHELM: A Holistic Evaluation… 33%, Aider code editing benchmark… 28% and 1 more
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The 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's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 Aider polyglot benchmark runs | 1 | 36% |
 AHELM: A Holistic Evaluation of Audio-Language Models (Stanford CRFM, arXiv:2508.21376v2) | 10 | 33% |
 Aider code editing benchmark runs (old edit leaderboard) | 1 | 28% |
| LiveCodeBench official website results (2025-08-01) | 1 | 2% |
| prior · pull toward the middle | 1% | |
| weighted average | = 56.3 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json