Model comparison
Gemini 3.5 Flash vs Gemini 3.8 Flash
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 54.2 on the Noometry Index.
Last verified . 43 shared benchmarks.
GoogleSummary
- They share 43 benchmarks with published results for both. Gemini 3.5 Flash scores higher in 1 category and Gemini 3.8 Flash in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Gemini 3.8 Flash leads 41.8 to 24.7.
- The biggest single-benchmark swing is APEX-Agents: 27.5% for Gemini 3.5 Flash and 64.3% for Gemini 3.8 Flash.
- Gemini 3.8 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $1.50 / $9 for Gemini 3.5 Flash.
Side by side
| Gemini 3.5 Flash | Gemini 3.8 Flash | |
|---|---|---|
| Provider | ||
| Noometry Index | 54.2 | 61.8 |
| Released | 2026-05-19 | 2026-09-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 66K | 66K |
| Input $ / M tokens | $1.50 | $0.75 |
| Output $ / M tokens | $9 | $3.75 |
| Results tracked | 54 | 50 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.8 Flash leads
Gemini 3.5 Flash: 49.4 (#49), Gemini 3.8 Flash: 59.2 (#15)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| DeepSWE | 37.4% | 73.8% |
| LMArena WebDev | 1499 | 1584 |
| SciCode | 53.1% | 56.6% |
| WeirdML | 62.6% | 84.8% |
| LMArena Coding | 1492 | 1510 |
| ALE-Bench | 911.02 | 1,270 |
| SWE-bench Verified | 79.3% | — |
| FrontierCode | — | 41.2% |
| CursorBench | — | 39.6% |
| FrontierSWE | — | 19.6% |
Agentic & Tool Use Gemini 3.8 Flash leads
Gemini 3.5 Flash: 24.7 (#114), Gemini 3.8 Flash: 41.8 (#21)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| APEX-Agents | 27.5% | 64.3% |
| GDP.pdf | 14% | 23.4% |
| Vending-Bench 2 | 5,396 | 5,094 |
| Remote Labor Index | — | 5.8% |
| GBAEval | 6.7% | — |
Reasoning Gemini 3.8 Flash leads
Gemini 3.5 Flash: 62.8 (#18), Gemini 3.8 Flash: 76.9 (#5)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| ARC-AGI-2 | 72.1% | 89.2% |
| NYT Connections (extended) | 92.6% | 97.4% |
| ARC-AGI-1 | 92.5% | 98.5% |
| CritPt | 13.1% | 18.3% |
| Chess Puzzles | 50% | 61% |
| LMArena Hard Prompts | 1488 | 1508 |
| Mystery Game Puzzles | 32% | 47% |
| DTBench | 94.7% | 95.7% |
| LMCA | 47.1% | 52.9% |
| Surface Evolver Bench | 58.1% | 76.9% |
| Epoch Capabilities Index | 154.46 | 156.71 |
| SimpleBench | 76.7% | — |
| EnigmaEval | 25.4% | — |
| EBR-Bench | 4.8% | — |
| ForecastBench | 59 | — |
Math Gemini 3.8 Flash leads
Gemini 3.5 Flash: 60.7 (#36), Gemini 3.8 Flash: 65.3 (#28)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 62.8% | 68.4% |
| FrontierMath Tier 4 | 26.8% | 22% |
| OTIS Mock AIME 2024-2025 | 95.6% | 98.9% |
| ProofBench | 31% | 48% |
| LMArena Math | 1504 | 1528 |
| MathArena Final-Answer Competitions | 76.3% | — |
| FrontierMath (Feb 2025 set) | 39% | — |
| FrontierMath Tier 4 (v1) | 14.6% | — |
Knowledge Gemini 3.8 Flash leads
Gemini 3.5 Flash: 66.3 (#11), Gemini 3.8 Flash: 74.8 (#2)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| GPQA Diamond | 92.8% | 95.4% |
| SimpleQA Verified | 66.2% | 69.7% |
| LMArena Expert | 1495 | 1524 |
| Humanity's Last Exam | — | 44.5% |
Multimodal Gemini 3.5 Flash leads
Gemini 3.5 Flash: 45.7 (#15), Gemini 3.8 Flash: 40.7 (#45)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Vision | 1310 | 1314 |
| Blueprint-Bench 2 | 33.6% | 38.6% |
| Furniture Assembly | — | 31.7% |
| LMArena Document | 1463 | — |
Multilingual Gemini 3.8 Flash leads
Gemini 3.5 Flash: 57.0 (#13), Gemini 3.8 Flash: 58.0 (#5)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Non-English | 1476 | 1491 |
| LMArena Chinese | 1526 | 1554 |
| LMArena French | 1490 | 1498 |
| LMArena German | 1492 | 1493 |
| LMArena Japanese | 1486 | 1502 |
| LMArena Korean | 1451 | 1459 |
| LMArena Russian | 1493 | 1515 |
| LMArena Spanish | 1480 | 1485 |
Instruction Following Gemini 3.8 Flash leads
Gemini 3.5 Flash: 77.0 (#30), Gemini 3.8 Flash: 78.0 (#13)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Instruction Following | 1467 | 1490 |
Long Context Too close to call
Gemini 3.5 Flash: 45.4 (#38), Gemini 3.8 Flash: 46.3 (#24)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Longer Query | 1482 | 1508 |
Writing & Preference Gemini 3.8 Flash leads
Gemini 3.5 Flash: 65.5 (#47), Gemini 3.8 Flash: 72.2 (#15)
| Benchmark | Gemini 3.5 Flash | Gemini 3.8 Flash |
|---|---|---|
| LMArena Text | 1482 | 1499 |
| LMArena Creative Writing | 1470 | 1492 |
| LMArena Multi-Turn | 1481 | 1501 |
| EQ-Bench Creative Writing | — | 1748 |
| EQ-Bench 4 | 1087 | — |
Frequently asked questions
Is Gemini 3.5 Flash better than Gemini 3.8 Flash?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 54.2 on the Noometry Index.
Which is cheaper, Gemini 3.5 Flash or Gemini 3.8 Flash?
Gemini 3.8 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; Gemini 3.5 Flash lists at $1.50 and $9.
Is Gemini 3.5 Flash or Gemini 3.8 Flash better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 49.4 in the Noometry coding category.
Which has the bigger context window?
Both accept 1.05M tokens.
How many benchmarks do Gemini 3.5 Flash and Gemini 3.8 Flash share?
43 benchmarks have published results for both models. Gemini 3.5 Flash has 54 scored results on Noometry and Gemini 3.8 Flash has 50.