Model comparison
Gemini 2.0 Flash-Lite vs GPT-4.1
Gemini 2.0 Flash-Lite is the stronger model overall, scoring 37.3 to 35.7 on the Noometry Index.
Last verified . 25 shared benchmarks.
GoogleSummary
- They share 25 benchmarks with published results for both. Gemini 2.0 Flash-Lite scores higher in 4 categories and GPT-4.1 in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemini 2.0 Flash-Lite leads 31.8 to 21.4.
- The biggest single-benchmark swing is GPQA (HELM): 50% for Gemini 2.0 Flash-Lite and 65.9% for GPT-4.1.
Side by side
| Gemini 2.0 Flash-Lite | GPT-4.1 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 37.3 | 35.7 |
| Released | 2025-02-05 | 2025-04-14 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 33K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $8 |
| Results tracked | 32 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 2.0 Flash-Lite leads
Gemini 2.0 Flash-Lite: 37.4 (#181), GPT-4.1: 34.2 (#228)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| LMArena Coding | 1322 | 1391 |
| SWE-bench Verified | — | 48.5% |
| SWE-bench Verified (bash only) | — | 39.6% |
| Aider Polyglot | — | 52.4% |
| WeirdML | — | 39% |
| LiveBench Coding | 47.1% | — |
| CadEval | — | 42% |
| ALE-Bench | — | 558.1 |
Agentic & Tool Use Not comparable
Gemini 2.0 Flash-Lite: —, GPT-4.1: 34.4 (#40)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 54% |
Reasoning Gemini 2.0 Flash-Lite leads
Gemini 2.0 Flash-Lite: 21.2 (#203), GPT-4.1: 11.7 (#334)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| LMArena Hard Prompts | 1324 | 1384 |
| DTBench | 52.5% | 68.3% |
| ForecastBench | 57.1 | 61.5 |
| ARC-AGI-2 | — | 0.4% |
| SimpleBench | — | 27% |
| Kagi LLM Benchmark | — | 52.3% |
| ARC-AGI-1 | — | 5.5% |
| Chess Puzzles | — | 6% |
| EnigmaEval | — | 2.2% |
| LiveBench Reasoning | 50.1% | — |
| LiveBench Data Analysis | 65.5% | — |
| LMCA | — | 25.6% |
| Epoch Capabilities Index | — | 136.79 |
| LiveBench | 54.3% | — |
Math Gemini 2.0 Flash-Lite leads
Gemini 2.0 Flash-Lite: 31.8 (#192), GPT-4.1: 21.4 (#269)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| Omni-MATH | 37.4% | 47.1% |
| LMArena Math | 1309 | 1369 |
| FrontierMath (Tiers 1-3) | — | 6% |
| OTIS Mock AIME 2024-2025 | — | 38.3% |
| LiveBench Math | 58.1% | — |
| MATH Level 5 | — | 83% |
| FrontierMath (Feb 2025 set) | — | 5.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge GPT-4.1 leads
Gemini 2.0 Flash-Lite: 34.6 (#185), GPT-4.1: 37.0 (#152)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| MMLU-Pro | 72% | 81.1% |
| GPQA (HELM) | 50% | 65.9% |
| LMArena Expert | 1305 | 1364 |
| GPQA Diamond | — | 66.9% |
| Humanity's Last Exam | — | 5.4% |
| SimpleQA Verified | — | 31.1% |
| Vectara Hallucination Rate | — | 5.6% |
Multimodal GPT-4.1 leads
Gemini 2.0 Flash-Lite: 30.6 (#103), GPT-4.1: 37.5 (#66)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| LMArena Vision | 1100 | 1210 |
| GeoBench | — | 72% |
Multilingual GPT-4.1 leads
Gemini 2.0 Flash-Lite: 46.2 (#156), GPT-4.1: 49.6 (#128)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1323 | 1370 |
| LMArena Chinese | 1339 | 1382 |
| LMArena French | 1346 | 1383 |
| LMArena German | 1306 | 1380 |
| LMArena Japanese | 1302 | 1320 |
| LMArena Korean | 1325 | 1339 |
| LMArena Russian | 1328 | 1377 |
| LMArena Spanish | 1313 | 1375 |
Instruction Following Too close to call
Gemini 2.0 Flash-Lite: 70.6 (#158), GPT-4.1: 71.5 (#148)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| IFEval | 82.4% | 83.8% |
| LMArena Instruction Following | 1305 | 1367 |
| LiveBench Instruction Following | 78.3% | — |
Long Context Too close to call
Gemini 2.0 Flash-Lite: 40.1 (#155), GPT-4.1: 40.0 (#158)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1321 | 1385 |
| Fiction.LiveBench | — | 63.9% |
Writing & Preference GPT-4.1 leads
Gemini 2.0 Flash-Lite: 51.9 (#172), GPT-4.1: 57.8 (#120)
| Benchmark | Gemini 2.0 Flash-Lite | GPT-4.1 |
|---|---|---|
| LMArena Text | 1330 | 1383 |
| LMArena Creative Writing | 1319 | 1363 |
| WildBench | 79% | 85.4% |
| LMArena Multi-Turn | 1307 | 1398 |
| EQ-Bench Creative Writing | — | 1420 |
| LiveBench Language | 34.3% | — |
Frequently asked questions
Is Gemini 2.0 Flash-Lite better than GPT-4.1?
Gemini 2.0 Flash-Lite is the stronger model overall, scoring 37.3 to 35.7 on the Noometry Index.
Is Gemini 2.0 Flash-Lite or GPT-4.1 better for coding?
Gemini 2.0 Flash-Lite scores higher on coding benchmarks: 37.4 versus 34.2 in the Noometry coding category.
How many benchmarks do Gemini 2.0 Flash-Lite and GPT-4.1 share?
25 benchmarks have published results for both models. Gemini 2.0 Flash-Lite has 32 scored results on Noometry and GPT-4.1 has 52.