whatmodel.sh / Gemini 2.0 Flash Thinking Experimental (01-21)
Gemini 2.0 Flash Thinking Experimental (01-21)
↗ share
+ contribute
Gemini 2.0 Flash Thinking Experimental (01-21) is a closed-weights language model from Google, released January 2025. Licence: proprietary.
Catalogue identity notes
Comparison model from the Epoch AI internal-evaluation CSVs (Epoch release date 2025-01-21). Experimental API model gemini-2.0-flash-thinking-exp-01-21; distinct from gemini-2.0-flash-001. Listed as shut down in the Gemini API changelog.
$ how it compares
The LLMs ranked either side of Gemini 2.0 Flash Thinking Experimental (01-21), out of 828 with a score, then more from Google.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 272 |  Ministral 3 14B Reasoning 2512 | 58±1 | local | ||
| 273 |  GPT-4o (2024-08-06) | 57±1 | api $4.38/M | ||
| 274 | Gemini 2.0 Flash Thinking Experimental (01-21) | 57±2 | comparison only | ||
| 275 |  DeepSeek R1 0528 Qwen3 8B | 57±1 | api $0.068/M +local | ||
| 276 |  Granite 4.2 3B | 57±1 | api $0.052/M +local | ||
| 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
agent tasks▁▁▂▆█▆▅▄▄▃▂▁ #246 of 340 * · 4.1% of its score · 1 test
Beats o1 (2024-12-17), o3-mini, DeepSeek R1 (Jan '25) here, though they rank higher overall. Trails SmolLM3 3B, Jamba 1.5 Mini, LFM2.5 1.2B Thinking here, though they rank lower overall.
coding▁▁▁▃▅▇█▇▅▄▃▁ #263 of 524 * · 45% of its score · 1 test
Beats Qwen3.5 4B, Qwen3-VL 32B Instruct, Solar Open 100B (102B-A12B) here, though they rank higher overall. Trails QwQ-32B-Preview, Ministral 3 3B Reasoning 2512, GPT-4 (0613) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Aider polyglot (225 exercises, two attempts) | 18.2% ✎ | ▂·▂▃▅▂▃█▆▄▂▂ beats 9/49 | 45% |
math▁▂▃▅▇▅▅█▄▁▁▁ #175 of 462 * · 17% of its score · 1 test
Beats MiMo-V2.5 (310B-A15B), MiniMax-M2.7, MiniMax-M2.1 here, though they rank higher overall. Trails Olmo 3 7B Think, Qwen3-VL 4B Thinking, Granite 4.2 3B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| OTIS mock AIME 2024–2025 | 57.8% ✎ | ▃▃▄▃▃▃▇▇█▃▂▅ beats 86/193 | 17% |
other tests1 not in a task · 34% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| GPQA Diamond | 57.1% ✎ | ▁▁▁▁▂▄▆▄▅▇█▅ beats 85/208 | 34% |
$ where to run it
No verified API or download recorded yet · add one
$ how the score was worked out · Epoch AI — internal evaluatio… 50%, Aider polyglot benchmark runs 45%, Berkeley Function Calling Lea… 4%
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 |
|---|---|---|
 Epoch AI — internal evaluations | 2 | 50% |
 Aider polyglot benchmark runs | 1 | 45% |
| Berkeley Function Calling Leaderboard result archive | 1 | 4% |
| prior · pull toward the middle | 1% | |
| weighted average | = 57.4 |
$ 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