Model comparison
Gemini 3.8 Flash vs GPT-3.5-turbo
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 2.0× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
GoogleSummary
- They share 27 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 8 categories and GPT-3.5-turbo in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 3.8 Flash leads 74.8 to 10.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.9% for Gemini 3.8 Flash and 2.2% for GPT-3.5-turbo.
- GPT-3.5-turbo is cheaper at $0.50 / $1.50 per million input/output tokens, against $0.75 / $3.75 for Gemini 3.8 Flash.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 16K.
Side by side
| Gemini 3.8 Flash | GPT-3.5-turbo | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 61.8 | 23.2 |
| Released | 2026-09-02 | 2023-03-01 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 16K |
| Max output | 66K | 4K |
| Input $ / M tokens | $0.75 | $0.50 |
| Output $ / M tokens | $3.75 | $1.50 |
| Results tracked | 50 | 44 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.8 Flash leads
Gemini 3.8 Flash: 59.2 (#15), GPT-3.5-turbo: 23.9 (#331)
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| WeirdML | 84.8% | 3.5% |
| LMArena Coding | 1510 | 1136 |
| DeepSWE | 73.8% | — |
| FrontierCode | 41.2% | — |
| CursorBench | 39.6% | — |
| LMArena WebDev | 1584 | — |
| FrontierSWE | 19.6% | — |
| SciCode | 56.6% | — |
| BigCodeBench Instruct | — | 39.1% |
| BigCodeBench Complete | — | 50.6% |
| ALE-Bench | 1,270 | — |
| HumanEval+ | — | 70.7% |
| MBPP+ | — | 69.7% |
Agentic & Tool Use Not comparable
Gemini 3.8 Flash: 41.8 (#21), GPT-3.5-turbo: —
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| APEX-Agents | 64.3% | — |
| Remote Labor Index | 5.8% | — |
| GDP.pdf | 23.4% | — |
| METR Time Horizons | — | 21.5% |
| Vending-Bench 2 | 5,094 | — |
Reasoning Gemini 3.8 Flash leads
Gemini 3.8 Flash: 76.9 (#5), GPT-3.5-turbo: 13.8 (#332)
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| Chess Puzzles | 61% | 0% |
| LMArena Hard Prompts | 1508 | 1108 |
| Mystery Game Puzzles | 47% | 3% |
| DTBench | 95.7% | 48.5% |
| LMCA | 52.9% | 9.7% |
| Epoch Capabilities Index | 156.71 | 118.55 |
| ARC-AGI-2 | 89.2% | — |
| NYT Connections (extended) | 97.4% | — |
| ARC-AGI-1 | 98.5% | — |
| CritPt | 18.3% | — |
| Surface Evolver Bench | 76.9% | — |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 61.6% |
| CommonsenseQA 2.0 | — | 57% |
| ForecastBench | — | 50.4 |
| WinoGrande | — | 81.6% |
Math Gemini 3.8 Flash leads
Gemini 3.8 Flash: 65.3 (#28), GPT-3.5-turbo: 6.3 (#327)
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| FrontierMath (Tiers 1-3) | 68.4% | 0% |
| OTIS Mock AIME 2024-2025 | 98.9% | 2.2% |
| LMArena Math | 1528 | 1142 |
| FrontierMath Tier 4 | 22% | — |
| ProofBench | 48% | — |
| MATH Level 5 | — | 15.9% |
| GSM8K | — | 57.8% |
Knowledge Gemini 3.8 Flash leads
Gemini 3.8 Flash: 74.8 (#2), GPT-3.5-turbo: 10.0 (#303)
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| GPQA Diamond | 95.4% | 28% |
| LMArena Expert | 1524 | 1070 |
| Humanity's Last Exam | 44.5% | — |
| SimpleQA Verified | 69.7% | — |
| ARC (AI2) Challenge | — | 87.4% |
| BoolQ | — | 87% |
| MMLU | — | 71.4% |
| OpenBookQA | — | 86% |
| TriviaQA | — | 85.8% |
Multimodal Not comparable
Gemini 3.8 Flash: 40.7 (#45), GPT-3.5-turbo: —
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| LMArena Vision | 1314 | — |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 31.7% | — |
Multilingual Gemini 3.8 Flash leads
Gemini 3.8 Flash: 58.0 (#5), GPT-3.5-turbo: 31.5 (#258)
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| LMArena Non-English | 1491 | 1108 |
| LMArena Chinese | 1554 | 1075 |
| LMArena French | 1498 | 1118 |
| LMArena German | 1493 | 1090 |
| LMArena Japanese | 1502 | 1043 |
| LMArena Korean | 1459 | 1019 |
| LMArena Russian | 1515 | 1123 |
| LMArena Spanish | 1485 | 1121 |
Instruction Following Gemini 3.8 Flash leads
Gemini 3.8 Flash: 78.0 (#13), GPT-3.5-turbo: 57.9 (#262)
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| LMArena Instruction Following | 1490 | 1119 |
Long Context Gemini 3.8 Flash leads
Gemini 3.8 Flash: 46.3 (#24), GPT-3.5-turbo: 34.0 (#254)
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| LMArena Longer Query | 1508 | 1121 |
Writing & Preference Gemini 3.8 Flash leads
Gemini 3.8 Flash: 72.2 (#15), GPT-3.5-turbo: 25.3 (#305)
| Benchmark | Gemini 3.8 Flash | GPT-3.5-turbo |
|---|---|---|
| LMArena Text | 1499 | 1125 |
| LMArena Creative Writing | 1492 | 1092 |
| EQ-Bench Creative Writing | 1748 | 451 |
| LMArena Multi-Turn | 1501 | 1117 |
Frequently asked questions
Is Gemini 3.8 Flash better than GPT-3.5-turbo?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 2.0× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.8 Flash or GPT-3.5-turbo?
GPT-3.5-turbo is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; Gemini 3.8 Flash lists at $0.75 and $3.75.
Is Gemini 3.8 Flash or GPT-3.5-turbo better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.8 Flash does, with 1.05M tokens against 16K.
How many benchmarks do Gemini 3.8 Flash and GPT-3.5-turbo share?
27 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and GPT-3.5-turbo has 44.