Model comparison
Gemini 2.5 Flash-Lite vs GPT-3.5-turbo
Gemini 2.5 Flash-Lite is the stronger model overall, scoring 37.0 to 23.2 on the Noometry Index.
Last verified . 21 shared benchmarks.
GoogleSummary
- They share 21 benchmarks with published results for both. Gemini 2.5 Flash-Lite scores higher in 7 categories and GPT-3.5-turbo in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemini 2.5 Flash-Lite leads 38.0 to 6.3.
- The biggest single-benchmark swing is WeirdML: 35.2% for Gemini 2.5 Flash-Lite and 3.5% for GPT-3.5-turbo.
- Gemini 2.5 Flash-Lite is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.50 / $1.50 for GPT-3.5-turbo.
- Gemini 2.5 Flash-Lite accepts more context: 1.05M tokens versus 16K.
Side by side
| Gemini 2.5 Flash-Lite | GPT-3.5-turbo | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 37.0 | 23.2 |
| Released | 2025-06-17 | 2023-03-01 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 16K |
| Max output | 66K | 4K |
| Input $ / M tokens | $0.10 | $0.50 |
| Output $ / M tokens | $0.40 | $1.50 |
| Results tracked | 33 | 44 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 2.5 Flash-Lite leads
Gemini 2.5 Flash-Lite: 38.5 (#173), GPT-3.5-turbo: 23.9 (#331)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| WeirdML | 35.2% | 3.5% |
| LMArena Coding | 1373 | 1136 |
| BigCodeBench Instruct | — | 39.1% |
| BigCodeBench Complete | — | 50.6% |
| ALE-Bench | 325.9 | — |
| HumanEval+ | — | 70.7% |
| MBPP+ | — | 69.7% |
Agentic & Tool Use Not comparable
Gemini 2.5 Flash-Lite: 28.0 (#96), GPT-3.5-turbo: —
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| Berkeley Function Calling Leaderboard | 36.9% | — |
| METR Time Horizons | — | 21.5% |
Reasoning Gemini 2.5 Flash-Lite leads
Gemini 2.5 Flash-Lite: 22.2 (#205), GPT-3.5-turbo: 13.8 (#332)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Hard Prompts | 1377 | 1108 |
| DTBench | 62.8% | 48.5% |
| LMCA | 18.1% | 9.7% |
| Epoch Capabilities Index | 133.94 | 118.55 |
| Kagi LLM Benchmark | 40.5% | — |
| Chess Puzzles | — | 0% |
| Mystery Game Puzzles | — | 3% |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 61.6% |
| CommonsenseQA 2.0 | — | 57% |
| ForecastBench | — | 50.4 |
| WinoGrande | — | 81.6% |
Math Gemini 2.5 Flash-Lite leads
Gemini 2.5 Flash-Lite: 38.0 (#144), GPT-3.5-turbo: 6.3 (#327)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Math | 1373 | 1142 |
| FrontierMath (Tiers 1-3) | — | 0% |
| OTIS Mock AIME 2024-2025 | — | 2.2% |
| Omni-MATH | 48% | — |
| MATH Level 5 | — | 15.9% |
| GSM8K | — | 57.8% |
Knowledge Gemini 2.5 Flash-Lite leads
Gemini 2.5 Flash-Lite: 32.5 (#210), GPT-3.5-turbo: 10.0 (#303)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Expert | 1373 | 1070 |
| GPQA Diamond | — | 28% |
| MMLU-Pro | 53.7% | — |
| Vectara Hallucination Rate | 3.3% | — |
| GPQA (HELM) | 30.9% | — |
| ARC (AI2) Challenge | — | 87.4% |
| BoolQ | — | 87% |
| MMLU | — | 71.4% |
| OpenBookQA | — | 86% |
| TriviaQA | — | 85.8% |
Multimodal Not comparable
Gemini 2.5 Flash-Lite: 29.1 (#114), GPT-3.5-turbo: —
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Vision | 1198 | — |
| VPCT | 30% | — |
Multilingual Gemini 2.5 Flash-Lite leads
Gemini 2.5 Flash-Lite: 49.3 (#134), GPT-3.5-turbo: 31.5 (#258)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Non-English | 1369 | 1108 |
| LMArena Chinese | 1404 | 1075 |
| LMArena French | 1388 | 1118 |
| LMArena German | 1389 | 1090 |
| LMArena Japanese | 1359 | 1043 |
| LMArena Korean | 1360 | 1019 |
| LMArena Russian | 1373 | 1123 |
| LMArena Spanish | 1396 | 1121 |
Instruction Following Gemini 2.5 Flash-Lite leads
Gemini 2.5 Flash-Lite: 70.0 (#168), GPT-3.5-turbo: 57.9 (#262)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Instruction Following | 1367 | 1119 |
| IFEval | 81% | — |
Long Context Too close to call
Gemini 2.5 Flash-Lite: 33.3 (#262), GPT-3.5-turbo: 34.0 (#254)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Longer Query | 1373 | 1121 |
| Fiction.LiveBench | 47.2% | — |
Writing & Preference Gemini 2.5 Flash-Lite leads
Gemini 2.5 Flash-Lite: 56.8 (#135), GPT-3.5-turbo: 25.3 (#305)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Text | 1379 | 1125 |
| LMArena Creative Writing | 1367 | 1092 |
| LMArena Multi-Turn | 1366 | 1117 |
| EQ-Bench Creative Writing | — | 451 |
| WildBench | 81.8% | — |
Frequently asked questions
Is Gemini 2.5 Flash-Lite better than GPT-3.5-turbo?
Gemini 2.5 Flash-Lite is the stronger model overall, scoring 37.0 to 23.2 on the Noometry Index.
Which is cheaper, Gemini 2.5 Flash-Lite or GPT-3.5-turbo?
Gemini 2.5 Flash-Lite is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; GPT-3.5-turbo lists at $0.50 and $1.50.
Is Gemini 2.5 Flash-Lite or GPT-3.5-turbo better for coding?
Gemini 2.5 Flash-Lite scores higher on coding benchmarks: 38.5 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash-Lite does, with 1.05M tokens against 16K.
How many benchmarks do Gemini 2.5 Flash-Lite and GPT-3.5-turbo share?
21 benchmarks have published results for both models. Gemini 2.5 Flash-Lite has 33 scored results on Noometry and GPT-3.5-turbo has 44.