Model comparison
Gemini 3.8 Flash vs Gemini 3 Pro
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 54.8 on the Noometry Index.
Last verified . 34 shared benchmarks.
GoogleSummary
- They share 34 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 9 categories and Gemini 3 Pro in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 52.5.
- The biggest single-benchmark swing is ARC-AGI-2: 89.2% for Gemini 3.8 Flash and 31.1% for Gemini 3 Pro.
Side by side
| Gemini 3.8 Flash | Gemini 3 Pro | |
|---|---|---|
| Provider | ||
| Noometry Index | 61.8 | 54.8 |
| Released | 2026-09-02 | 2025-11-18 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.75 | — |
| Output $ / M tokens | $3.75 | — |
| Results tracked | 50 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.8 Flash leads
Gemini 3.8 Flash: 59.2 (#15), Gemini 3 Pro: 51.6 (#39)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena WebDev | 1584 | 1440 |
| WeirdML | 84.8% | 69.9% |
| LMArena Coding | 1510 | 1481 |
| ALE-Bench | 1,270 | 1,177 |
| SWE-bench Verified | — | 72.9% |
| DeepSWE | 73.8% | — |
| FrontierCode | 41.2% | — |
| SWE-bench Verified (bash only) | — | 74.2% |
| CursorBench | 39.6% | — |
| SWE-bench Multilingual | — | 68.7% |
| FrontierSWE | 19.6% | — |
| SciCode | 56.6% | — |
| GSO | — | 18.6% |
| AlgoTune | — | 1.83 |
Agentic & Tool Use Gemini 3.8 Flash leads
Gemini 3.8 Flash: 41.8 (#21), Gemini 3 Pro: 40.6 (#23)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| Remote Labor Index | 5.8% | 1.3% |
| Vending-Bench 2 | 5,094 | 5,478 |
| Terminal-Bench | — | 69.4% |
| APEX-Agents | 64.3% | — |
| Berkeley Function Calling Leaderboard | — | 72.5% |
| GDPval | — | 40.3% |
| τ²-bench Airline | — | 80.5% |
| τ²-bench Banking | — | 18% |
| τ²-bench Retail | — | 75.9% |
| τ²-bench Telecom | — | 91% |
| DeepResearch Bench | — | 46.3% |
| BALROG | — | 58.1% |
| GDP.pdf | 23.4% | — |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 71% |
Reasoning Gemini 3.8 Flash leads
Gemini 3.8 Flash: 76.9 (#5), Gemini 3 Pro: 52.5 (#31)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| ARC-AGI-2 | 89.2% | 31.1% |
| NYT Connections (extended) | 97.4% | 94.4% |
| ARC-AGI-1 | 98.5% | 75% |
| CritPt | 18.3% | 6.9% |
| Chess Puzzles | 61% | 31% |
| LMArena Hard Prompts | 1508 | 1480 |
| Epoch Capabilities Index | 156.71 | 152.92 |
| SimpleBench | — | 76.4% |
| Kagi LLM Benchmark | — | 80.1% |
| EnigmaEval | — | 18.2% |
| Mystery Game Puzzles | 47% | — |
| DTBench | 95.7% | — |
| LMCA | 52.9% | — |
| Surface Evolver Bench | 76.9% | — |
| ForecastBench | — | 61.2 |
Math Gemini 3.8 Flash leads
Gemini 3.8 Flash: 65.3 (#28), Gemini 3 Pro: 49.9 (#59)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 91.4% |
| ProofBench | 48% | 20% |
| LMArena Math | 1528 | 1476 |
| FrontierMath (Tiers 1-3) | 68.4% | — |
| FrontierMath Tier 4 | 22% | — |
| MathArena Final-Answer Competitions | — | 67% |
| Omni-MATH | — | 55.5% |
| FrontierMath (Feb 2025 set) | — | 37.6% |
| FrontierMath Tier 4 (v1) | — | 18.8% |
Knowledge Gemini 3.8 Flash leads
Gemini 3.8 Flash: 74.8 (#2), Gemini 3 Pro: 64.4 (#16)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| GPQA Diamond | 95.4% | 92.6% |
| Humanity's Last Exam | 44.5% | 37.5% |
| LMArena Expert | 1524 | 1475 |
| SimpleQA Verified | 69.7% | — |
| MMLU-Pro | — | 90.3% |
| Vectara Hallucination Rate | — | 13.6% |
| GPQA (HELM) | — | 80.3% |
Multimodal Gemini 3 Pro leads
Gemini 3.8 Flash: 40.7 (#45), Gemini 3 Pro: 57.6 (#2)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena Vision | 1314 | 1305 |
| GeoBench | — | 84% |
| VPCT | — | 91% |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 31.7% | — |
| LMArena Document | — | 1434 |
Multilingual Gemini 3.8 Flash leads
Gemini 3.8 Flash: 58.0 (#5), Gemini 3 Pro: 56.9 (#16)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena Non-English | 1491 | 1474 |
| LMArena Chinese | 1554 | 1523 |
| LMArena French | 1498 | 1492 |
| LMArena German | 1493 | 1515 |
| LMArena Japanese | 1502 | 1510 |
| LMArena Korean | 1459 | 1448 |
| LMArena Russian | 1515 | 1493 |
| LMArena Spanish | 1485 | 1470 |
Instruction Following Gemini 3.8 Flash leads
Gemini 3.8 Flash: 78.0 (#13), Gemini 3 Pro: 76.3 (#45)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena Instruction Following | 1490 | 1458 |
| IFEval | — | 87.7% |
Long Context Gemini 3.8 Flash leads
Gemini 3.8 Flash: 46.3 (#24), Gemini 3 Pro: 44.0 (#79)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena Longer Query | 1508 | 1471 |
| CL-bench | — | 15.8% |
Writing & Preference Gemini 3.8 Flash leads
Gemini 3.8 Flash: 72.2 (#15), Gemini 3 Pro: 66.4 (#35)
| Benchmark | Gemini 3.8 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena Text | 1499 | 1479 |
| LMArena Creative Writing | 1492 | 1482 |
| EQ-Bench Creative Writing | 1748 | 1525 |
| LMArena Multi-Turn | 1501 | 1484 |
| WildBench | — | 85.9% |
Frequently asked questions
Is Gemini 3.8 Flash better than Gemini 3 Pro?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 54.8 on the Noometry Index.
Is Gemini 3.8 Flash or Gemini 3 Pro better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 51.6 in the Noometry coding category.
How many benchmarks do Gemini 3.8 Flash and Gemini 3 Pro share?
34 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and Gemini 3 Pro has 67.