Model comparison
Gemini 3.5 Flash vs GPT-5.4 Pro
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 54.2 on the Noometry Index. Gemini 3.5 Flash costs 20× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
GoogleSummary
- They share 14 benchmarks with published results for both. Gemini 3.5 Flash scores higher in 1 category and GPT-5.4 Pro in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 Pro leads 72.4 to 60.7.
- The biggest single-benchmark swing is FrontierMath Tier 4: 26.8% for Gemini 3.5 Flash and 58.5% for GPT-5.4 Pro.
- Gemini 3.5 Flash is cheaper at $1.50 / $9 per million input/output tokens, against $30 / $180 for GPT-5.4 Pro.
- GPT-5.4 Pro accepts more context: 1.05M tokens versus 1.05M.
Side by side
| Gemini 3.5 Flash | GPT-5.4 Pro | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 54.2 | 58.9 |
| Released | 2026-05-19 | 2026-03-05 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 66K | 128K |
| Input $ / M tokens | $1.50 | $30 |
| Output $ / M tokens | $9 | $180 |
| Results tracked | 54 | 16 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.5 Flash leads
Gemini 3.5 Flash: 49.4 (#49), GPT-5.4 Pro: 43.2 (#90)
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| WeirdML | 62.6% | 57.4% |
| SWE-bench Verified | 79.3% | — |
| DeepSWE | 37.4% | — |
| LMArena WebDev | 1499 | — |
| SciCode | 53.1% | — |
| LMArena Coding | 1492 | — |
| ALE-Bench | 911.02 | — |
Agentic & Tool Use Not comparable
Gemini 3.5 Flash: 24.7 (#114), GPT-5.4 Pro: —
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| APEX-Agents | 27.5% | — |
| GBAEval | 6.7% | — |
| GDP.pdf | 14% | — |
| Vending-Bench 2 | 5,396 | — |
Reasoning GPT-5.4 Pro leads
Gemini 3.5 Flash: 62.8 (#18), GPT-5.4 Pro: 70.7 (#13)
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| ARC-AGI-2 | 72.1% | 83.3% |
| SimpleBench | 76.7% | 74.1% |
| ARC-AGI-1 | 92.5% | 94.5% |
| CritPt | 13.1% | 30% |
| Chess Puzzles | 50% | 58.6% |
| EnigmaEval | 25.4% | 23.8% |
| Epoch Capabilities Index | 154.46 | 158.93 |
| NYT Connections (extended) | 92.6% | — |
| EBR-Bench | 4.8% | — |
| LMArena Hard Prompts | 1488 | — |
| Mystery Game Puzzles | 32% | — |
| DTBench | 94.7% | — |
| LMCA | 47.1% | — |
| Surface Evolver Bench | 58.1% | — |
| ForecastBench | 59 | — |
Math GPT-5.4 Pro leads
Gemini 3.5 Flash: 60.7 (#36), GPT-5.4 Pro: 72.4 (#22)
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | 62.8% | 82.5% |
| FrontierMath Tier 4 | 26.8% | 58.5% |
| FrontierMath (Feb 2025 set) | 39% | 50% |
| FrontierMath Tier 4 (v1) | 14.6% | 37.5% |
| MathArena Final-Answer Competitions | 76.3% | — |
| OTIS Mock AIME 2024-2025 | 95.6% | — |
| ProofBench | 31% | — |
| LMArena Math | 1504 | — |
Knowledge GPT-5.4 Pro leads
Gemini 3.5 Flash: 66.3 (#11), GPT-5.4 Pro: 68.3 (#7)
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| GPQA Diamond | 92.8% | 94.6% |
| SimpleQA Verified | 66.2% | 46.3% |
| Humanity's Last Exam | — | 44.3% |
| Vectara Hallucination Rate | — | 8.3% |
| LMArena Expert | 1495 | — |
Multimodal Not comparable
Gemini 3.5 Flash: 45.7 (#15), GPT-5.4 Pro: —
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Vision | 1310 | — |
| Blueprint-Bench 2 | 33.6% | — |
| LMArena Document | 1463 | — |
Multilingual Not comparable
Gemini 3.5 Flash: 57.0 (#13), GPT-5.4 Pro: —
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Non-English | 1476 | — |
| LMArena Chinese | 1526 | — |
| LMArena French | 1490 | — |
| LMArena German | 1492 | — |
| LMArena Japanese | 1486 | — |
| LMArena Korean | 1451 | — |
| LMArena Russian | 1493 | — |
| LMArena Spanish | 1480 | — |
Instruction Following Not comparable
Gemini 3.5 Flash: 77.0 (#30), GPT-5.4 Pro: —
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Instruction Following | 1467 | — |
Long Context Not comparable
Gemini 3.5 Flash: 45.4 (#38), GPT-5.4 Pro: —
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Longer Query | 1482 | — |
Writing & Preference Not comparable
Gemini 3.5 Flash: 65.5 (#47), GPT-5.4 Pro: —
| Benchmark | Gemini 3.5 Flash | GPT-5.4 Pro |
|---|---|---|
| LMArena Text | 1482 | — |
| LMArena Creative Writing | 1470 | — |
| EQ-Bench 4 | 1087 | — |
| LMArena Multi-Turn | 1481 | — |
Frequently asked questions
Is Gemini 3.5 Flash better than GPT-5.4 Pro?
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 54.2 on the Noometry Index. Gemini 3.5 Flash costs 20× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.5 Flash or GPT-5.4 Pro?
Gemini 3.5 Flash is cheaper. It lists at $1.50 per million input tokens and $9 per million output tokens; GPT-5.4 Pro lists at $30 and $180.
Is Gemini 3.5 Flash or GPT-5.4 Pro better for coding?
Gemini 3.5 Flash scores higher on coding benchmarks: 49.4 versus 43.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 Pro does, with 1.05M tokens against 1.05M.
How many benchmarks do Gemini 3.5 Flash and GPT-5.4 Pro share?
14 benchmarks have published results for both models. Gemini 3.5 Flash has 54 scored results on Noometry and GPT-5.4 Pro has 16.