Model comparison
Gemini 3.1 Flash Lite vs GPT-5.2
GPT-5.2 is the stronger model overall, scoring 54.1 to 40.8 on the Noometry Index. Gemini 3.1 Flash Lite costs 8.6× less per token, which makes it the better buy when GPT-5.2's lead doesn't matter for your workload.
Last verified . 35 shared benchmarks.
GoogleSummary
- They share 35 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 0 categories and GPT-5.2 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.2 leads 50.2 to 22.9.
- The biggest single-benchmark swing is NYT Connections (extended): 8.2% for Gemini 3.1 Flash Lite and 83.6% for GPT-5.2.
- Gemini 3.1 Flash Lite is cheaper at $0.25 / $1.50 per million input/output tokens, against $1.75 / $14 for GPT-5.2.
- Gemini 3.1 Flash Lite accepts more context: 1.05M tokens versus 400K.
Side by side
| Gemini 3.1 Flash Lite | GPT-5.2 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 40.8 | 54.1 |
| Released | 2026-03-03 | 2025-12-11 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 400K |
| Max output | 66K | 128K |
| Input $ / M tokens | $0.25 | $1.75 |
| Output $ / M tokens | $1.50 | $14 |
| Results tracked | 38 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 leads
Gemini 3.1 Flash Lite: 37.8 (#188), GPT-5.2: 51.6 (#37)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| LMArena WebDev | 1256 | 1416 |
| WeirdML | 52.2% | 72.2% |
| LMArena Coding | 1400 | 1447 |
| ALE-Bench | 797.73 | 1,294 |
| SWE-bench Verified | — | 73.8% |
| SWE-bench Verified (bash only) | — | 72.8% |
| SWE-bench Multilingual | — | 66.7% |
| SciCode | 41.9% | — |
| GSO | — | 27.4% |
| AlgoTune | — | 2.05 |
Agentic & Tool Use GPT-5.2 leads
Gemini 3.1 Flash Lite: 30.2 (#79), GPT-5.2: 40.2 (#24)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| DeepResearch Bench | 37.3% | 41.1% |
| Terminal-Bench | — | 64.9% |
| Berkeley Function Calling Leaderboard | — | 55.9% |
| GDPval | — | 49.7% |
| Remote Labor Index | — | 2.5% |
| τ²-bench Airline | — | 83% |
| τ²-bench Banking | — | 32.2% |
| τ²-bench Retail | — | 81.6% |
| τ²-bench Telecom | — | 89.7% |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 75.3% |
| Vending-Bench 2 | — | 3,591 |
Reasoning GPT-5.2 leads
Gemini 3.1 Flash Lite: 22.9 (#186), GPT-5.2: 50.2 (#35)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| Kagi LLM Benchmark | 67.2% | 73.3% |
| NYT Connections (extended) | 8.2% | 83.6% |
| Chess Puzzles | 25% | 49% |
| EnigmaEval | 3% | 10.4% |
| LMArena Hard Prompts | 1407 | 1445 |
| DTBench | 76.8% | 90.9% |
| LMCA | 35% | 43.9% |
| Epoch Capabilities Index | 144.47 | 153.45 |
| ForecastBench | 54.4 | 60.1 |
| ARC-AGI-2 | — | 52.9% |
| SimpleBench | — | 45.8% |
| ARC-AGI-1 | — | 86.2% |
| CritPt | 1.1% | — |
| Thematic Generalization | 63.3% | — |
| EBR-Bench | — | 23% |
| Mystery Game Puzzles | — | 23% |
Math GPT-5.2 leads
Gemini 3.1 Flash Lite: 40.7 (#90), GPT-5.2: 60.0 (#38)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 27.7% | 67.4% |
| OTIS Mock AIME 2024-2025 | 80% | 96.1% |
| LMArena Math | 1428 | 1440 |
| FrontierMath Tier 4 | — | 31.7% |
| MathArena Final-Answer Competitions | — | 72% |
| ProofBench | — | 15% |
| FrontierMath (Feb 2025 set) | — | 40.7% |
| FrontierMath Tier 4 (v1) | — | 18.8% |
Knowledge GPT-5.2 leads
Gemini 3.1 Flash Lite: 41.9 (#104), GPT-5.2: 59.3 (#32)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| GPQA Diamond | 81.8% | 91.4% |
| Humanity's Last Exam | 8.6% | 27.8% |
| Vectara Hallucination Rate | 8.2% | 8.4% |
| LMArena Expert | 1398 | 1445 |
| SimpleQA Verified | — | 37.1% |
Multimodal GPT-5.2 leads
Gemini 3.1 Flash Lite: 39.4 (#60), GPT-5.2: 51.3 (#7)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| LMArena Vision | 1240 | 1268 |
| VPCT | — | 84% |
| Furniture Assembly | — | 38.3% |
| LMArena Document | — | 1405 |
Multilingual GPT-5.2 leads
Gemini 3.1 Flash Lite: 52.3 (#86), GPT-5.2: 53.4 (#67)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| LMArena Non-English | 1411 | 1425 |
| LMArena Chinese | 1461 | 1460 |
| LMArena French | 1424 | 1455 |
| LMArena German | 1429 | 1448 |
| LMArena Japanese | 1413 | 1420 |
| LMArena Korean | 1392 | 1392 |
| LMArena Russian | 1420 | 1440 |
| LMArena Spanish | 1421 | 1433 |
Instruction Following GPT-5.2 leads
Gemini 3.1 Flash Lite: 72.7 (#131), GPT-5.2: 74.7 (#89)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| LMArena Instruction Following | 1377 | 1417 |
Long Context GPT-5.2 leads
Gemini 3.1 Flash Lite: 42.5 (#122), GPT-5.2: 44.0 (#78)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| LMArena Longer Query | 1394 | 1428 |
| CL-bench | — | 18.2% |
Writing & Preference GPT-5.2 leads
Gemini 3.1 Flash Lite: 60.9 (#94), GPT-5.2: 66.8 (#32)
| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 |
|---|---|---|
| LMArena Text | 1416 | 1439 |
| LMArena Creative Writing | 1401 | 1401 |
| LMArena Multi-Turn | 1417 | 1458 |
| EQ-Bench Creative Writing | — | 1703 |
Frequently asked questions
Is Gemini 3.1 Flash Lite better than GPT-5.2?
GPT-5.2 is the stronger model overall, scoring 54.1 to 40.8 on the Noometry Index. Gemini 3.1 Flash Lite costs 8.6× less per token, which makes it the better buy when GPT-5.2's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.1 Flash Lite or GPT-5.2?
Gemini 3.1 Flash Lite is cheaper. It lists at $0.25 per million input tokens and $1.50 per million output tokens; GPT-5.2 lists at $1.75 and $14.
Is Gemini 3.1 Flash Lite or GPT-5.2 better for coding?
GPT-5.2 scores higher on coding benchmarks: 51.6 versus 37.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.1 Flash Lite does, with 1.05M tokens against 400K.
How many benchmarks do Gemini 3.1 Flash Lite and GPT-5.2 share?
35 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and GPT-5.2 has 67.