Model comparison
Claude Opus 5 vs Gemini 3.8 Flash
Claude Opus 5 is the stronger model overall, scoring 67.8 to 61.8 on the Noometry Index. Gemini 3.8 Flash costs 6.7× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Last verified . 47 shared benchmarks.
GoogleSummary
- They share 47 benchmarks with published results for both. Claude Opus 5 scores higher in 9 categories and Gemini 3.8 Flash in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 5 leads 86.2 to 65.3.
- The biggest single-benchmark swing is FrontierMath Tier 4: 73.2% for Claude Opus 5 and 22% for Gemini 3.8 Flash.
- Gemini 3.8 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $5 / $25 for Claude Opus 5.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 1M.
Side by side
| Claude Opus 5 | Gemini 3.8 Flash | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 67.8 | 61.8 |
| Released | 2026-07-24 | 2026-09-02 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 128K | 66K |
| Input $ / M tokens | $5 | $0.75 |
| Output $ / M tokens | $25 | $3.75 |
| Results tracked | 57 | 50 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 5 leads
Claude Opus 5: 67.5 (#5), Gemini 3.8 Flash: 59.2 (#15)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| DeepSWE | 73.6% | 73.8% |
| FrontierCode | 53.4% | 41.2% |
| CursorBench | 46.6% | 39.6% |
| LMArena WebDev | 1691 | 1584 |
| FrontierSWE | 52% | 19.6% |
| SciCode | 56.4% | 56.6% |
| WeirdML | 91.8% | 84.8% |
| LMArena Coding | 1534 | 1510 |
| ALE-Bench | 2,165 | 1,270 |
Agentic & Tool Use Claude Opus 5 leads
Claude Opus 5: 55.6 (#1), Gemini 3.8 Flash: 41.8 (#21)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| APEX-Agents | 65.8% | 64.3% |
| GDP.pdf | 24% | 23.4% |
| Vending-Bench 2 | 11,182 | 5,094 |
| OSWorld 2.0 | 31.4% | — |
| Remote Labor Index | — | 5.8% |
| τ²-bench Banking | 48.7% | — |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |
Reasoning Too close to call
Claude Opus 5: 77.2 (#4), Gemini 3.8 Flash: 76.9 (#5)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| ARC-AGI-2 | 90.4% | 89.2% |
| NYT Connections (extended) | 94.3% | 97.4% |
| ARC-AGI-1 | 97.5% | 98.5% |
| CritPt | 29.1% | 18.3% |
| Chess Puzzles | 42% | 61% |
| LMArena Hard Prompts | 1526 | 1508 |
| Mystery Game Puzzles | 59% | 47% |
| DTBench | 97.9% | 95.7% |
| LMCA | 64.5% | 52.9% |
| Epoch Capabilities Index | 162.78 | 156.71 |
| SimpleBench | 80.6% | — |
| EBR-Bench | 45.7% | — |
| Surface Evolver Bench | — | 76.9% |
| Bench to the Future 3 | 0.12 | — |
Math Claude Opus 5 leads
Claude Opus 5: 86.2 (#8), Gemini 3.8 Flash: 65.3 (#28)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 85.6% | 68.4% |
| FrontierMath Tier 4 | 73.2% | 22% |
| OTIS Mock AIME 2024-2025 | 98.9% | 98.9% |
| ProofBench | 99% | 48% |
| LMArena Math | 1531 | 1528 |
Knowledge Gemini 3.8 Flash leads
Claude Opus 5: 66.8 (#9), Gemini 3.8 Flash: 74.8 (#2)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| GPQA Diamond | 93.9% | 95.4% |
| SimpleQA Verified | 59.9% | 69.7% |
| LMArena Expert | 1557 | 1524 |
| Humanity's Last Exam | — | 44.5% |
Multimodal Claude Opus 5 leads
Claude Opus 5: 50.8 (#8), Gemini 3.8 Flash: 40.7 (#45)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Vision | 1319 | 1314 |
| Blueprint-Bench 2 | 30.4% | 38.6% |
| Furniture Assembly | 60.8% | 31.7% |
| LMArena Document | 1516 | — |
Multilingual Too close to call
Claude Opus 5: 58.8 (#4), Gemini 3.8 Flash: 58.0 (#5)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Non-English | 1501 | 1491 |
| LMArena Chinese | 1574 | 1554 |
| LMArena French | 1519 | 1498 |
| LMArena German | 1524 | 1493 |
| LMArena Japanese | 1516 | 1502 |
| LMArena Korean | 1521 | 1459 |
| LMArena Russian | 1507 | 1515 |
| LMArena Spanish | 1519 | 1485 |
Instruction Following Claude Opus 5 leads
Claude Opus 5: 79.2 (#7), Gemini 3.8 Flash: 78.0 (#13)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Instruction Following | 1517 | 1490 |
Long Context Too close to call
Claude Opus 5: 46.5 (#21), Gemini 3.8 Flash: 46.3 (#24)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Longer Query | 1515 | 1508 |
Writing & Preference Claude Opus 5 leads
Claude Opus 5: 79.2 (#1), Gemini 3.8 Flash: 72.2 (#15)
| Benchmark | Claude Opus 5 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Text | 1507 | 1499 |
| LMArena Creative Writing | 1491 | 1492 |
| EQ-Bench Creative Writing | 2133 | 1748 |
| LMArena Multi-Turn | 1499 | 1501 |
| EQ-Bench 4 | 1385 | — |
Frequently asked questions
Is Claude Opus 5 better than Gemini 3.8 Flash?
Claude Opus 5 is the stronger model overall, scoring 67.8 to 61.8 on the Noometry Index. Gemini 3.8 Flash costs 6.7× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 5 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; Claude Opus 5 lists at $5 and $25.
Is Claude Opus 5 or Gemini 3.8 Flash better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 59.2 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.8 Flash does, with 1.05M tokens against 1M.
How many benchmarks do Claude Opus 5 and Gemini 3.8 Flash share?
47 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Gemini 3.8 Flash has 50.