Gemini 3.5 Flash Lite vs GPT-5.2
Updated October 21, 2026. 68 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick GPT-5.2 for coding: it scores 58.9 on the Vibe Coding Index against 33.6 for Gemini 3.5 Flash Lite.
Pick Gemini 3.5 Flash Lite when output tokens cost 5.6x less, or when a solved task costs $0.41 against $1.91.
- Shared results
- 68
- Gemini 3.5 Flash Lite only
- 15
- GPT-5.2 only
- 43
- Like-for-like categories
- 6 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, four public coding benchmarks
GPT-5.2
GPT-5.2 scores 58.9 against 33.6.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-5.2
GPT-5.2 leads, 30.4 against 22.2.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Gemini 3.5 Flash Lite has a AA Agentic Index result, 14.3.
Directional only
Reasoning
GPQA Diamond
GPT-5.2
GPT-5.2 leads, 91.7% against 83.8%.
Same board, both measured
Math
AIME
No call
Only GPT-5.2 has a AIME result, 96.9%.
Directional only
Knowledge
MMLU Pro
GPT-5.2
GPT-5.2 leads, 86.2% against 85.8%.
Same board, both measured
Web design
Design Arena: websites
No call
Only GPT-5.2 has a Design Arena: websites result, 1196.
Directional only
Long documents
Listed context window
Gemini 3.5 Flash Lite
Gemini 3.5 Flash Lite takes 1.05M tokens against 400K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Gemini 3.5 Flash Lite
Gemini 3.5 Flash Lite $0.0015, GPT-5.2 $0.0088 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Gemini 3.5 Flash Lite
Gemini 3.5 Flash Lite $0.41, GPT-5.2 $1.91.
Published attempt cost and pass rate
At a glance
- Coding index
- Gemini 3.5 Flash Lite 33.6 from 1 benchmark, GPT-5.2 58.9 from 2 benchmarks.
- Widest category
- Math, 17.1 points apart on average over 6 shared benchmarks.
- Widest single gap
- ARC-AGI-2: Gemini 3.5 Flash Lite 10.3% against GPT-5.2 52.9%, 42.6 points apart on a 0 to 100 scale.
- Output price
- Gemini 3.5 Flash Lite $2.50 against GPT-5.2 $14.00 per million output tokens, 5.6x apart.
- Context window
- Gemini 3.5 Flash Lite takes 1.05M tokens, GPT-5.2 400K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Gemini 3.5 Flash Lite | Bars | GPT-5.2 | Ties |
|---|---|---|---|---|
| Coding | 3 | 5 | ||
| Intelligence | 0 | 2 | ||
| Reasoning | 4 | 10 | ||
| Math | 0 | 6 | ||
| Knowledge | 6 | 4 | ||
| General | 22 | 6 | ||
| All shared boards | 35 | 33 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index35
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.2 | Leader |
|---|---|---|---|
| SWE-bench Verified | Not measured | 72.8% | Only GPT-5.2 measured |
| SWE-bench Multilingual | Not measured | 66.7% | Only GPT-5.2 measured |
| LMArena WebDev | 1435 | 1417 | Gemini 3.5 Flash Lite |
| AA Coding Index | 49.3 | Not measured | Only Gemini 3.5 Flash Lite measured |
| LiveCodeBench (Vals AI) | 79.0% | 85.4% | GPT-5.2 |
| SWE-bench Verified (Vals AI) | 75.0% | 75.8% | GPT-5.2 |
| Vibe Code Bench (Vals AI) | 37.2% | 53.5% | GPT-5.2 |
| Code migration (Vals AI) | 6.2% | Not measured | Only Gemini 3.5 Flash Lite measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only Gemini 3.5 Flash Lite measured |
| Terminal-Bench 2.1 (Vals AI) | 50.2% | Not measured | Only Gemini 3.5 Flash Lite measured |
| CyberBench (Vals AI) | 59.1% | Not measured | Only Gemini 3.5 Flash Lite measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 89.4% | Only GPT-5.2 measured |
| SciCode | 41.3% | Not measured | Only Gemini 3.5 Flash Lite measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 47.0% | Only GPT-5.2 measured |
| Terminal-Bench | Not measured | 64.9% | Only GPT-5.2 measured |
| GSO-Bench | Not measured | 27.5% | Only GPT-5.2 measured |
| WeirdML | 39.0% | 72.2% | GPT-5.2 |
| ALE-Bench | 765.3 | 1293.5 | GPT-5.2 |
| LMArena Text: coding | 1453 | 1442 | Gemini 3.5 Flash Lite |
| LMArena Text: software and IT | 1451 | 1434 | Gemini 3.5 Flash Lite |
| LMArena WebDev: React | 1408 | Not measured | Only Gemini 3.5 Flash Lite measured |
| LMArena WebDev: HTML | Not measured | 1427 | Only GPT-5.2 measured |
| LMArena WebDev: from an image | 1448 | Not measured | Only Gemini 3.5 Flash Lite measured |
| Neither model has a result on the other 15 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.2 | Leader |
|---|---|---|---|
| AA Intelligence Index | 22.2 | 30.4 | GPT-5.2 |
| Epoch Capabilities Index | 144.9 | 153.7 | GPT-5.2 |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index17 boards, none shared
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.2 | Leader |
|---|---|---|---|
| AA Agentic Index | 14.3 | Not measured | Only Gemini 3.5 Flash Lite measured |
| Design Arena: agents building web apps | Not measured | 1086 | Only GPT-5.2 measured |
| Design Arena: agents building full-stack apps | Not measured | 1043 | Only GPT-5.2 measured |
| Design Arena: agents building mobile apps | Not measured | 1094 | Only GPT-5.2 measured |
| Design Arena: agents building Android apps | Not measured | 1064 | Only GPT-5.2 measured |
| Design Arena: agents building Godot games | Not measured | 1142 | Only GPT-5.2 measured |
| Finance Agent (Vals AI) | 47.4% | Not measured | Only Gemini 3.5 Flash Lite measured |
| Tax Agent Bench (Vals AI) | 7.2% | Not measured | Only Gemini 3.5 Flash Lite measured |
| Legal Research Bench (Vals AI) | 13.9% | Not measured | Only Gemini 3.5 Flash Lite measured |
| Excel Modeling Benchmark (Vals AI) | 43.2% | Not measured | Only Gemini 3.5 Flash Lite measured |
| τ²-Bench Telecom | Not measured | 84.8% | Only GPT-5.2 measured |
| APEX-Agents | 29.3% | Not measured | Only Gemini 3.5 Flash Lite measured |
| METR time horizons | Not measured | 75.3% | Only GPT-5.2 measured |
| Remote Labor Index | Not measured | 2.5% | Only GPT-5.2 measured |
| GDPval | Not measured | 49.7% | Only GPT-5.2 measured |
| DeepResearch Bench | Not measured | 41.1% | Only GPT-5.2 measured |
| Vending-Bench 2 | Not measured | $3,591 | Only GPT-5.2 measured |
| Neither model has a result on the other 22 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index410
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.2 | Leader |
|---|---|---|---|
| GPQA Diamond | 83.8% | 91.7% | GPT-5.2 |
| MMMU Pro (Vals AI) | 83.6% | 86.7% | GPT-5.2 |
| GPQA Diamond (Artificial Analysis run) | 83.8% | 90.3% | GPT-5.2 |
| Humanity's Last Exam (Artificial Analysis run) | 18.8% | 37.7% | GPT-5.2 |
| GPQA Diamond (Epoch AI run) | 83.3% | 91.4% | GPT-5.2 |
| ARC-AGI | 53.5% | 86.2% | GPT-5.2 |
| ARC-AGI-2 | 10.3% | 52.9% | GPT-5.2 |
| SimpleBench | Not measured | 45.8% | Only GPT-5.2 measured |
| Chess puzzles | 21.0% | 48.0% | GPT-5.2 |
| Mystery game puzzles | 19.0% | 23.0% | GPT-5.2 |
| EBR-bench | Not measured | 23.0% | Only GPT-5.2 measured |
| VPCT | Not measured | 84.0% | Only GPT-5.2 measured |
| LMArena Text: hard prompts | 1441 | 1428 | Gemini 3.5 Flash Lite |
| LMArena Text: expert | 1434 | 1445 | GPT-5.2 |
| LMArena Text: hard prompts in English | 1440 | 1431 | Gemini 3.5 Flash Lite |
| LMArena Vision: diagrams | 1284 | 1265 | Gemini 3.5 Flash Lite |
| LMArena Vision: homework | 1291 | 1276 | Gemini 3.5 Flash Lite |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.2 | Leader |
|---|---|---|---|
| AIME | Not measured | 96.9% | Only GPT-5.2 measured |
| MGSM (Vals AI) | Not measured | 94.0% | Only GPT-5.2 measured |
| AA Math Index | Not measured | 99.0 | Only GPT-5.2 measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 99.0% | Only GPT-5.2 measured |
| OTIS Mock AIME | 71.1% | 96.1% | GPT-5.2 |
| FrontierMath, tiers 1 to 3 | 26.0% | 67.4% | GPT-5.2 |
| FrontierMath, tier 4 | 0.0% | 31.7% | GPT-5.2 |
| ProofBench | 13.0% | 15.0% | GPT-5.2 |
| LMArena Text: math | 1434 | 1440 | GPT-5.2 |
| LMArena Text: mathematical work | 1433 | 1442 | GPT-5.2 |
| Neither model has a result on the other 3 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index64
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.2 | Leader |
|---|---|---|---|
| MMLU Pro | 85.8% | 86.2% | GPT-5.2 |
| LegalBench (Vals AI) | 84.1% | 82.8% | Gemini 3.5 Flash Lite |
| CaseLaw (Vals AI) | Not measured | 66.0% | Only GPT-5.2 measured |
| CorpFin (Vals AI) | 60.7% | 65.9% | GPT-5.2 |
| TaxEval (Vals AI) | 72.6% | 75.8% | GPT-5.2 |
| MortgageTax (Vals AI) | 68.7% | 67.1% | Gemini 3.5 Flash Lite |
| MedQA (Vals AI) | Not measured | 94.1% | Only GPT-5.2 measured |
| MedCode (Vals AI) | 43.5% | 49.8% | GPT-5.2 |
| MMLU Pro (Artificial Analysis run) | Not measured | 87.4% | Only GPT-5.2 measured |
| SimpleQA Verified | Not measured | 37.1% | Only GPT-5.2 measured |
| LMArena Text: science | 1451 | 1425 | Gemini 3.5 Flash Lite |
| LMArena Text: medicine and healthcare | 1438 | 1425 | Gemini 3.5 Flash Lite |
| LMArena Text: legal and government | 1431 | 1427 | Gemini 3.5 Flash Lite |
| LMArena Text: business and finance | 1423 | 1414 | Gemini 3.5 Flash Lite |
| Neither model has a result on the other 3 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.2 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1196 | Only GPT-5.2 measured |
| Design Arena: UI components | Not measured | 1190 | Only GPT-5.2 measured |
| Design Arena: game dev | Not measured | 1196 | Only GPT-5.2 measured |
| Design Arena: data viz | Not measured | 1202 | Only GPT-5.2 measured |
| Design Arena: 3D | Not measured | 1079 | Only GPT-5.2 measured |
| Design Arena: all code categories | Not measured | 1171 | Only GPT-5.2 measured |
| Design Arena: SVG | Not measured | 1137 | Only GPT-5.2 measured |
| Design Arena: ASCII art | Not measured | 1153 | Only GPT-5.2 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index226
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.2 | Leader |
|---|---|---|---|
| Public Benefits Bench (Vals AI) | 51.8% | Not measured | Only Gemini 3.5 Flash Lite measured |
| SAGE (Vals AI) | 47.3% | 49.3% | GPT-5.2 |
| MedScribe (Vals AI) | 70.9% | 84.4% | GPT-5.2 |
| IFBench | Not measured | 75.4% | Only GPT-5.2 measured |
| Long context reasoning (AA-LCR) | 76.0% | 82.7% | GPT-5.2 |
| DTBench | 83.5% | 90.9% | GPT-5.2 |
| LMCA | 37.6% | 43.9% | GPT-5.2 |
| CL-bench | Not measured | 18.2% | Only GPT-5.2 measured |
| Furniture assembly | Not measured | 38.3% | Only GPT-5.2 measured |
| LMArena Text | 1435 | 1416 | Gemini 3.5 Flash Lite |
| LMArena Text: instruction following | 1419 | 1409 | Gemini 3.5 Flash Lite |
| LMArena Text: creative writing | 1421 | 1376 | Gemini 3.5 Flash Lite |
| LMArena Vision | 1269 | 1246 | Gemini 3.5 Flash Lite |
| LMArena Text: English | 1438 | 1422 | Gemini 3.5 Flash Lite |
| LMArena Text: non-English | 1427 | 1405 | Gemini 3.5 Flash Lite |
| LMArena Text: Chinese | 1468 | 1447 | Gemini 3.5 Flash Lite |
| LMArena Text: German | 1454 | 1444 | Gemini 3.5 Flash Lite |
| LMArena Text: Japanese | 1426 | 1420 | Gemini 3.5 Flash Lite |
| LMArena Text: Korean | 1407 | 1374 | Gemini 3.5 Flash Lite |
| LMArena Text: French | 1451 | 1454 | GPT-5.2 |
| LMArena Text: Spanish | 1441 | 1413 | Gemini 3.5 Flash Lite |
| LMArena Text: Russian | 1443 | 1415 | Gemini 3.5 Flash Lite |
| LMArena Text: Polish | 1437 | 1426 | Gemini 3.5 Flash Lite |
| LMArena Text: multi-turn | 1445 | 1419 | Gemini 3.5 Flash Lite |
| LMArena Text: longer queries | 1435 | 1413 | Gemini 3.5 Flash Lite |
| LMArena Text: writing and language | 1428 | 1392 | Gemini 3.5 Flash Lite |
| LMArena Text: entertainment and media | 1414 | 1378 | Gemini 3.5 Flash Lite |
| LMArena Vision: English | 1269 | 1243 | Gemini 3.5 Flash Lite |
| LMArena Vision: Chinese | 1311 | 1305 | Gemini 3.5 Flash Lite |
| LMArena Vision: reading text | 1280 | 1257 | Gemini 3.5 Flash Lite |
| LMArena Vision: humor | 1287 | 1227 | Gemini 3.5 Flash Lite |
| LMArena Vision: captioning | Not measured | 1267 | Only GPT-5.2 measured |
| LMArena Vision: recognizing things | Not measured | 1232 | Only GPT-5.2 measured |
| LMArena Vision: creative writing | 1260 | 1225 | Gemini 3.5 Flash Lite |
| LMArena Document | Not measured | 1405 | Only GPT-5.2 measured |
| Neither model has a result on the other 11 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.2 | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Gemini 3.5 Flash Lite costs less
Repository review
50K input + 3K output tokens
Gemini 3.5 Flash Lite costs less
Agent loop
200K input + 10K output tokens
Gemini 3.5 Flash Lite costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Google Gemini 3.5 Flash Lite | OpenAI GPT-5.2 |
|---|---|---|
| Maker | OpenAI | |
| Weights | No public weights listed | No public weights listed |
| Released | July 21, 2026 | December 10, 2025 |
| Context window | 1.05M | 400K |
| Input $/Mtok | $0.30 | $1.75 |
| Output $/Mtok | $2.50 | $14.00 |
| Coding index | 33.6 | 58.9 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | $0.41 | $1.91 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Gemini 3.5 Flash Lite better than GPT-5.2 for coding?
Pick GPT-5.2 for coding: it scores 58.9 on the Vibe Coding Index against 33.6 for Gemini 3.5 Flash Lite. Pick Gemini 3.5 Flash Lite when output tokens cost 5.6x less, or when a solved task costs $0.41 against $1.91.
Which is cheaper, Gemini 3.5 Flash Lite or GPT-5.2?
Gemini 3.5 Flash Lite costs $0.30 per million input tokens and $2.50 per million output tokens; GPT-5.2 costs $1.75 and $14.00. Per solved SWE-bench Verified task, Gemini 3.5 Flash Lite costs $0.41 and GPT-5.2 $1.91.
Which has the bigger context window, Gemini 3.5 Flash Lite or GPT-5.2?
Gemini 3.5 Flash Lite, with 1.05M tokens against 400K.
How many benchmarks is the Gemini 3.5 Flash Lite vs GPT-5.2 comparison based on?
68 benchmarks have a result for both models as of October 21, 2026. Gemini 3.5 Flash Lite's coding index rests on 1 independent benchmark and GPT-5.2's on 2. Where only one model has a result, the page says so instead of naming a winner.