Model comparison
Gemini 2.5 Flash-Lite vs GPT-5
GPT-5 is the stronger model overall, scoring 50.9 to 37.0 on the Noometry Index. Gemini 2.5 Flash-Lite costs 20× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
GoogleSummary
- They share 32 benchmarks with published results for both. Gemini 2.5 Flash-Lite scores higher in 0 categories and GPT-5 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in long context, where GPT-5 leads 69.5 to 33.3.
- The biggest single-benchmark swing is Fiction.LiveBench: 47.2% for Gemini 2.5 Flash-Lite and 97.2% for GPT-5.
- Gemini 2.5 Flash-Lite is cheaper at $0.10 / $0.40 per million input/output tokens, against $1.25 / $10 for GPT-5.
- Gemini 2.5 Flash-Lite accepts more context: 1.05M tokens versus 400K.
Side by side
| Gemini 2.5 Flash-Lite | GPT-5 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 37.0 | 50.9 |
| Released | 2025-06-17 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 400K |
| Max output | 66K | 128K |
| Input $ / M tokens | $0.10 | $1.25 |
| Output $ / M tokens | $0.40 | $10 |
| Results tracked | 33 | 69 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 leads
Gemini 2.5 Flash-Lite: 38.5 (#173), GPT-5: 50.3 (#47)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| WeirdML | 35.2% | 60.7% |
| LMArena Coding | 1373 | 1436 |
| ALE-Bench | 325.9 | 1,162 |
| SWE-bench Verified | — | 73.6% |
| SWE-bench Verified (bash only) | — | 65% |
| Aider Polyglot | — | 88% |
| LMArena WebDev | — | 1418 |
| SciCode | — | 42.9% |
| GSO | — | 6.9% |
| AlgoTune | — | 1.67 |
Agentic & Tool Use GPT-5 leads
Gemini 2.5 Flash-Lite: 28.0 (#96), GPT-5: 33.1 (#56)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| Terminal-Bench | — | 49.6% |
| Berkeley Function Calling Leaderboard | 36.9% | — |
| GDPval | — | 34.8% |
| Remote Labor Index | — | 1.7% |
| DeepResearch Bench | — | 49.6% |
| BALROG | — | 32.8% |
| LMArena Search | — | 1133 |
| METR Time Horizons | — | 69.6% |
Reasoning GPT-5 leads
Gemini 2.5 Flash-Lite: 22.2 (#205), GPT-5: 38.3 (#64)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| Kagi LLM Benchmark | 40.5% | 72.7% |
| LMArena Hard Prompts | 1377 | 1416 |
| DTBench | 62.8% | 90.7% |
| LMCA | 18.1% | 40% |
| Epoch Capabilities Index | 133.94 | 150 |
| ARC-AGI-2 | — | 9.9% |
| SimpleBench | — | 56.7% |
| ARC-AGI-1 | — | 65.7% |
| CritPt | — | 12.6% |
| Chess Puzzles | — | 37% |
| EnigmaEval | — | 10.5% |
| EBR-Bench | — | 12.7% |
| Mystery Game Puzzles | — | 23% |
| ForecastBench | — | 61.4 |
Math GPT-5 leads
Gemini 2.5 Flash-Lite: 38.0 (#144), GPT-5: 55.0 (#44)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| Omni-MATH | 48% | 64.7% |
| LMArena Math | 1373 | 1407 |
| FrontierMath (Tiers 1-3) | — | 55.4% |
| FrontierMath Tier 4 | — | 22% |
| OTIS Mock AIME 2024-2025 | — | 91.4% |
| ProofBench | — | 18% |
| MATH Level 5 | — | 98.1% |
| FrontierMath (Feb 2025 set) | — | 32.4% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge GPT-5 leads
Gemini 2.5 Flash-Lite: 32.5 (#210), GPT-5: 56.6 (#43)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| MMLU-Pro | 53.7% | 86.3% |
| Vectara Hallucination Rate | 3.3% | 14.7% |
| GPQA (HELM) | 30.9% | 79.2% |
| LMArena Expert | 1373 | 1419 |
| GPQA Diamond | — | 86.2% |
| Humanity's Last Exam | — | 25.3% |
| SimpleQA Verified | — | 50.1% |
| Confabulations | — | 10.3% |
Multimodal GPT-5 leads
Gemini 2.5 Flash-Lite: 29.1 (#114), GPT-5: 46.8 (#13)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| LMArena Vision | 1198 | 1232 |
| VPCT | 30% | 66% |
| GeoBench | — | 81% |
Multilingual GPT-5 leads
Gemini 2.5 Flash-Lite: 49.3 (#134), GPT-5: 51.4 (#110)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| LMArena Non-English | 1369 | 1397 |
| LMArena Chinese | 1404 | 1422 |
| LMArena French | 1388 | 1410 |
| LMArena German | 1389 | 1416 |
| LMArena Japanese | 1359 | 1409 |
| LMArena Korean | 1360 | 1360 |
| LMArena Russian | 1373 | 1406 |
| LMArena Spanish | 1396 | 1399 |
Instruction Following GPT-5 leads
Gemini 2.5 Flash-Lite: 70.0 (#168), GPT-5: 73.8 (#113)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| IFEval | 81% | 87.5% |
| LMArena Instruction Following | 1367 | 1388 |
Long Context GPT-5 leads
Gemini 2.5 Flash-Lite: 33.3 (#262), GPT-5: 69.5 (#2)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| Fiction.LiveBench | 47.2% | 97.2% |
| LMArena Longer Query | 1373 | 1399 |
Writing & Preference GPT-5 leads
Gemini 2.5 Flash-Lite: 56.8 (#135), GPT-5: 63.4 (#65)
| Benchmark | Gemini 2.5 Flash-Lite | GPT-5 |
|---|---|---|
| LMArena Text | 1379 | 1406 |
| LMArena Creative Writing | 1367 | 1365 |
| WildBench | 81.8% | 85.7% |
| LMArena Multi-Turn | 1366 | 1426 |
| Short-Story Creative Writing | — | 86% |
| EQ-Bench Creative Writing | — | 1627 |
Frequently asked questions
Is Gemini 2.5 Flash-Lite better than GPT-5?
GPT-5 is the stronger model overall, scoring 50.9 to 37.0 on the Noometry Index. Gemini 2.5 Flash-Lite costs 20× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.
Which is cheaper, Gemini 2.5 Flash-Lite or GPT-5?
Gemini 2.5 Flash-Lite is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; GPT-5 lists at $1.25 and $10.
Is Gemini 2.5 Flash-Lite or GPT-5 better for coding?
GPT-5 scores higher on coding benchmarks: 50.3 versus 38.5 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash-Lite does, with 1.05M tokens against 400K.
How many benchmarks do Gemini 2.5 Flash-Lite and GPT-5 share?
32 benchmarks have published results for both models. Gemini 2.5 Flash-Lite has 33 scored results on Noometry and GPT-5 has 69.