Model comparison
Gemini 2.5 Flash vs o1
o1 is the stronger model overall, scoring 40.9 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 31× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Last verified . 35 shared benchmarks.
GoogleSummary
- They share 35 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 5 categories and o1 in 5 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where o1 leads 46.1 to 35.8.
- The biggest single-benchmark swing is VPCT: 46.2% for Gemini 2.5 Flash and 37% for o1.
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $15 / $60 for o1.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 200K.
Side by side
| Gemini 2.5 Flash | o1 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 39.3 | 40.9 |
| Released | 2025-04-17 | 2024-09-12 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 200K |
| Max output | 66K | 100K |
| Input $ / M tokens | $0.30 | $15 |
| Output $ / M tokens | $2.50 | $60 |
| Results tracked | 54 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
Gemini 2.5 Flash: 35.8 (#220), o1: 46.1 (#70)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| Aider Polyglot | 55.1% | 61.7% |
| WeirdML | 41.9% | 47.6% |
| LMArena Coding | 1424 | 1367 |
| SWE-bench Verified (bash only) | 28.7% | — |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| ALE-Bench | 661.88 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use Gemini 2.5 Flash leads
Gemini 2.5 Flash: 30.8 (#74), o1: 24.6 (#117)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| Terminal-Bench | 17.1% | — |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| Cybench | — | 10% |
| BALROG | 33.5% | — |
| METR Time Horizons | — | 51.1% |
| Vending-Bench 2 | 548.84 | — |
Reasoning o1 leads
Gemini 2.5 Flash: 18.1 (#286), o1: 27.9 (#111)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| SimpleBench | 41.2% | 41.7% |
| ARC-AGI-1 | 33.3% | 30.7% |
| EnigmaEval | 2.7% | 5.7% |
| LMArena Hard Prompts | 1422 | 1371 |
| DTBench | 76.5% | 74.7% |
| LMCA | 27.5% | 22.3% |
| Epoch Capabilities Index | 143.03 | 141.91 |
| ARC-AGI-2 | 2.5% | — |
| Kagi LLM Benchmark | 56.8% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 15% |
| LiveBench Reasoning | — | 91.6% |
| LiveBench Data Analysis | — | 65.5% |
| ForecastBench | 60.6 | — |
| LiveBench | — | 75.7% |
Math Gemini 2.5 Flash leads
Gemini 2.5 Flash: 39.9 (#98), o1: 36.1 (#175)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 73.3% |
| LMArena Math | 1415 | 1388 |
| FrontierMath (Feb 2025 set) | 4.8% | 9.3% |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| Omni-MATH | 38.5% | — |
| LiveBench Math | — | 80.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge o1 leads
Gemini 2.5 Flash: 36.4 (#168), o1: 41.5 (#110)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| Humanity's Last Exam | 12.1% | 8% |
| Confabulations | 16.8% | 11.7% |
| LMArena Expert | 1426 | 1361 |
| GPQA Diamond | — | 76.8% |
| SimpleQA Verified | — | 41.1% |
| MMLU-Pro | 63.9% | — |
| Vectara Hallucination Rate | 7.8% | — |
| GPQA (HELM) | 39% | — |
Multimodal Gemini 2.5 Flash leads
Gemini 2.5 Flash: 41.8 (#32), o1: 34.2 (#93)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| LMArena Vision | 1253 | 1168 |
| GeoBench | 76% | 80% |
| VPCT | 46.2% | 37% |
| SpatialViz-Bench | 36.9% | 41.4% |
Multilingual Gemini 2.5 Flash leads
Gemini 2.5 Flash: 52.3 (#88), o1: 48.6 (#142)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| LMArena Non-English | 1409 | 1358 |
| LMArena Chinese | 1450 | 1394 |
| LMArena French | 1433 | 1344 |
| LMArena German | 1418 | 1337 |
| LMArena Japanese | 1405 | 1346 |
| LMArena Korean | 1385 | 1396 |
| LMArena Russian | 1415 | 1356 |
| LMArena Spanish | 1421 | 1345 |
Instruction Following Too close to call
Gemini 2.5 Flash: 75.7 (#54), o1: 74.8 (#86)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| LMArena Instruction Following | 1405 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
| IFEval | 89.8% | — |
Long Context o1 leads
Gemini 2.5 Flash: 47.5 (#17), o1: 50.3 (#9)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| Fiction.LiveBench | 77.8% | 83.3% |
| LMArena Longer Query | 1419 | 1378 |
Writing & Preference o1 leads
Gemini 2.5 Flash: 53.8 (#157), o1: 55.6 (#144)
| Benchmark | Gemini 2.5 Flash | o1 |
|---|---|---|
| LMArena Text | 1417 | 1366 |
| LMArena Creative Writing | 1400 | 1348 |
| Short-Story Creative Writing | 76.5% | 70.2% |
| LMArena Multi-Turn | 1408 | 1369 |
| EQ-Bench Creative Writing | 1137 | — |
| WildBench | 81.7% | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is Gemini 2.5 Flash better than o1?
o1 is the stronger model overall, scoring 40.9 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 31× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Which is cheaper, Gemini 2.5 Flash or o1?
Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; o1 lists at $15 and $60.
Is Gemini 2.5 Flash or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 35.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 200K.
How many benchmarks do Gemini 2.5 Flash and o1 share?
35 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and o1 has 52.