Model comparison
Gemini 3.5 Flash Lite vs GPT-5.1
GPT-5.1 is the stronger model overall, scoring 49.0 to 41.5 on the Noometry Index. Gemini 3.5 Flash Lite costs 4.0× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
GoogleSummary
- They share 32 benchmarks with published results for both. Gemini 3.5 Flash Lite scores higher in 0 categories and GPT-5.1 in 10 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.1 leads 52.2 to 25.9.
- The biggest single-benchmark swing is WeirdML: 39% for Gemini 3.5 Flash Lite and 60.8% for GPT-5.1.
- Gemini 3.5 Flash Lite is cheaper at $0.30 / $2.50 per million input/output tokens, against $1.25 / $10 for GPT-5.1.
- Gemini 3.5 Flash Lite accepts more context: 1.05M tokens versus 400K.
Side by side
| Gemini 3.5 Flash Lite | GPT-5.1 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 41.5 | 49.0 |
| Released | 2026-07-21 | 2025-11-13 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 400K |
| Max output | 66K | 128K |
| Input $ / M tokens | $0.30 | $1.25 |
| Output $ / M tokens | $2.50 | $10 |
| Results tracked | 39 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
Gemini 3.5 Flash Lite: 42.4 (#104), GPT-5.1: 46.4 (#66)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| LMArena WebDev | 1440 | 1395 |
| SciCode | 41.3% | 43.3% |
| WeirdML | 39% | 60.8% |
| LMArena Coding | 1453 | 1454 |
| ALE-Bench | 765.27 | 1,192 |
| SWE-bench Verified | — | 68% |
| SWE-bench Verified (bash only) | — | 66% |
| GSO | — | 13.7% |
| LiveBench Coding | — | 72.5% |
Agentic & Tool Use GPT-5.1 leads
Gemini 3.5 Flash Lite: 25.7 (#107), GPT-5.1: 32.7 (#60)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| Terminal-Bench | — | 47.6% |
| APEX-Agents | 29.3% | — |
| DeepResearch Bench | — | 42.8% |
| GDP.pdf | 10% | — |
| LMArena Search | — | 1199 |
| Vending-Bench 2 | — | 1,473 |
Reasoning GPT-5.1 leads
Gemini 3.5 Flash Lite: 27.8 (#114), GPT-5.1: 39.8 (#58)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| ARC-AGI-2 | 10.3% | 17.6% |
| ARC-AGI-1 | 53.5% | 72.8% |
| CritPt | 0% | 4.9% |
| Chess Puzzles | 22% | 32% |
| LMArena Hard Prompts | 1441 | 1457 |
| Mystery Game Puzzles | 19% | 19% |
| DTBench | 83.5% | 90.1% |
| LMCA | 37.6% | 43.9% |
| Epoch Capabilities Index | 145.13 | 149.64 |
| SimpleBench | — | 53.2% |
| NYT Connections (extended) | 60.4% | — |
| EnigmaEval | — | 11.2% |
| LiveBench Reasoning | — | 95.8% |
| LiveBench Data Analysis | — | 72.1% |
| ForecastBench | — | 58.1 |
| LiveBench | — | 78.8% |
Math GPT-5.1 leads
Gemini 3.5 Flash Lite: 25.9 (#264), GPT-5.1: 52.2 (#51)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 88.6% |
| LMArena Math | 1433 | 1447 |
| FrontierMath (Tiers 1-3) | 26% | — |
| FrontierMath Tier 4 | 0% | — |
| ProofBench | 13% | — |
| Omni-MATH | — | 46.4% |
| LiveBench Math | — | 94.5% |
| FrontierMath (Feb 2025 set) | — | 31% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge Too close to call
Gemini 3.5 Flash Lite: 50.1 (#72), GPT-5.1: 50.6 (#71)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 83.3% | 87.6% |
| LMArena Expert | 1435 | 1470 |
| Humanity's Last Exam | — | 23.7% |
| SimpleQA Verified | — | 48% |
| MMLU-Pro | — | 57.9% |
| Vectara Hallucination Rate | — | 10.9% |
| GPQA (HELM) | — | 44.2% |
Multimodal GPT-5.1 leads
Gemini 3.5 Flash Lite: 41.1 (#40), GPT-5.1: 44.8 (#19)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1268 | 1250 |
| VPCT | — | 58.7% |
| LMArena Document | — | 1403 |
Multilingual Too close to call
Gemini 3.5 Flash Lite: 53.5 (#63), GPT-5.1: 53.8 (#56)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1427 | 1431 |
| LMArena Chinese | 1469 | 1495 |
| LMArena French | 1451 | 1450 |
| LMArena German | 1454 | 1438 |
| LMArena Japanese | 1427 | 1453 |
| LMArena Korean | 1407 | 1401 |
| LMArena Russian | 1443 | 1435 |
| LMArena Spanish | 1442 | 1433 |
Instruction Following GPT-5.1 leads
Gemini 3.5 Flash Lite: 74.8 (#85), GPT-5.1: 83.9 (#1)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| LMArena Instruction Following | 1420 | 1443 |
| LiveBench Instruction Following | — | 93.3% |
| IFEval | — | 93.5% |
Long Context GPT-5.1 leads
Gemini 3.5 Flash Lite: 43.9 (#84), GPT-5.1: 47.6 (#14)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1435 | 1447 |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |
Writing & Preference Too close to call
Gemini 3.5 Flash Lite: 64.2 (#57), GPT-5.1: 64.5 (#55)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.1 |
|---|---|---|
| LMArena Text | 1435 | 1443 |
| LMArena Creative Writing | 1420 | 1427 |
| LMArena Multi-Turn | 1445 | 1450 |
| EQ-Bench Creative Writing | 1559 | — |
| WildBench | — | 86.3% |
| LiveBench Language | — | 80.2% |
Frequently asked questions
Is Gemini 3.5 Flash Lite better than GPT-5.1?
GPT-5.1 is the stronger model overall, scoring 49.0 to 41.5 on the Noometry Index. Gemini 3.5 Flash Lite costs 4.0× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.5 Flash Lite or GPT-5.1?
Gemini 3.5 Flash Lite is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; GPT-5.1 lists at $1.25 and $10.
Is Gemini 3.5 Flash Lite or GPT-5.1 better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.4 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.5 Flash Lite does, with 1.05M tokens against 400K.
How many benchmarks do Gemini 3.5 Flash Lite and GPT-5.1 share?
32 benchmarks have published results for both models. Gemini 3.5 Flash Lite has 39 scored results on Noometry and GPT-5.1 has 63.