whatmodel.sh / Gemini 3.1 Flash-Lite
Gemini 3.1 Flash-Lite
↗ share
+ contribute
Gemini 3.1 Flash-Lite is a closed-weights language model from Google, released March 2026.
$ how it compares
The LLMs ranked either side of Gemini 3.1 Flash-Lite, out of 828 with a score, then more from Google.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 132 |  Step 3.7 Flash (198B-A11B) | 67±1 | api $0.44/M +local | ||
| 133 |  Gemma 4 31B | 67±1 | api $0.19/M +local | ||
| 134 | Gemini 3.1 Flash-Lite | 67±1 | api $0.56/M | ||
| 135 |  Qwen3-Coder-Next (80B-A3B) | 67±2 | api $0.60/M +local | ||
| 136 | Gemini 2.5 Pro Preview (03-25) | 67±2 | comparison only | ||
| 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▁▁▃▆█▆▅▅▅▄▃▁ #130 of 300 * · 4.5% of its score · 1 test
Beats o3, Gemini 2.5 Pro here, though they rank higher overall. Trails SmolLM3 3B, Jamba 1.5 Large, GPT-4o (2024-05-13) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| K-BrowseComp-Verified (Korean web-browsing agent questions, pass@1) | 11.3% ✎ | ▃····█··▃█·▃ beats 2/8 | 4.5% |
coding▁▁▁▃▆██▇▆▄▃▁ #156 of 458 * · 15% of its score · 1 test
Beats Gemini 2.5 Pro here, though they rank higher overall. Trails QwQ-32B-Preview, Ministral 3 8B Reasoning 2512, Phi-4-reasoning 14B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Arena WebDev (Code Arena) | 1255.7 ✎ | ▁▂▄▃▅█▄▇▆▄▁▁ beats 15/119 | 15% |
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #76 of 393 * · 8.2% of its score · 1 test
Beats LongCat 2.0 (1.6T-A48B), Mistral Large 4 (public preview), Inkling Small (266B-A12B) here, though they rank higher overall. Trails Gemma 4 12B, Gemma 4 26B-A4B, Nemotron Cascade 2 30B-A3B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena (text) | 1432.7 ✎ | ▁▁▁▂▃▃▃▆▆▇█▃ beats 288/379 | 8.2% |
math▁▂▃▅▆▅▄█▄▁▁▁ #115 of 398 · 39% of its score · 3 tests
Beats Claude Sonnet 4.6, MiniMax-M3, MiMo-V2.5 (310B-A15B) here, though they rank higher overall. Trails Olmo 3.1 32B Think, Gemini 2.5 Flash, Qwen3-VL 32B Thinking here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| FrontierMath tiers 1–3 v2 | 27.7% ✎ | ▂··▁▁▃▄▄█▅▃▃ beats 19/82 | 31% |
| OTIS mock AIME 2024–2025 | 80.0% ✎ | ▃▃▄▃▃▃▇▇█▃▂▅ beats 115/193 | 6.0% |
| MathConstraint full release, no tools (329 instances) | 24.9% ✎ | ▄··▄▄·▄▄█▄·▄ beats 2/8 | 1.8% |
vision & documents▁▁▃▂▁▃▄▂▆▇█▄ #60 of 147 * · 18% of its score · 2 tests
Beats DeepSeek V4 Flash Vision (Exp), GPT-5, Inkling Small (266B-A12B) here, though they rank higher overall. Trails Gemma 4 26B-A4B, Qwen3.6 Flash, Qwen3.5 35B-A3B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| CharXiv (reasoning questions) lab | 73.2% ✎ | ▃·▃·▃▃▃·▃▃█▅ beats 3/11 | 12% |
| Arena vision (style controlled) | 1234.6 ✎ | ▁▁▂▂▂▂▄▅▅▆██ beats 91/145 | 5.7% |
web search▂▃▂▆▄█▃▃▁▁·▁ #75 of 80 * · 4.5% of its score · 1 test
Trails GPT-4.1 nano, GPT-4o mini (2024-07-18), Mistral Small 3.1 24B (2503) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| K-BrowseComp-Verified (Korean web-browsing agent questions, pass@1) | 11.3% ✎ | ▃····█··▃█·▃ beats 2/8 | 4.5% |
other tests3 not in a task · 16% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| GPQA Diamond | 81.8% ✎ | ▁▁▁▁▂▄▆▄▅▇█▅ beats 130/208 | 12% |
| Chess Puzzles (Epoch) | 20.0% ✎ | █▁▁▁▁▂▁▄▂▂▁▁ beats 82/135 | 3.2% |
| Vectara hallucination leaderboard (HHEM-2.3, factual consistency rate) | 91.8% ✎ | ▁▂▁▂▅█▃▃▄▃▁▁ beats 61/98 | 0.0% |
$ where to run it
Through an API
api · $0.56/M · $0.25 in / $1.5 out per 1M tokens · 1M context · ai.google.dev ↗ · price sourceStandard base token rate from models.dev. Provider-specific tiers, long-context surcharges, caching and promotions may differ.
$ how the score was worked out · Epoch AI — internal evaluatio… 53%, LMArena leaderboard dataset 29%, Google DeepMind Gemini model… 12% and 3 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 |
|---|---|---|
 Epoch AI — internal evaluations | 4 | 53% |
 LMArena leaderboard dataset | 3 | 29% |
 Google DeepMind Gemini model page and model cards (lab-reported scores) · lab-reported | 1 | 12% |
 K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts (arXiv:2606.02404v1) | 1 | 5% |
| MathConstraint: Automated Generation of Verified Combinatorial Reasoning Instances for LLMs — authors' own model runs | 1 | 2% |
 Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository) | 1 | 0% |
| prior · pull toward the middle | 0% | |
| weighted average | = 66.9 |
$ 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