Model comparison
Claude Opus 5 vs Gemini 3.5 Flash
Claude Opus 5 is the stronger model overall, scoring 67.8 to 54.2 on the Noometry Index. Gemini 3.5 Flash costs 3.0× 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 10 categories and Gemini 3.5 Flash in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude Opus 5 leads 55.6 to 24.7.
- The biggest single-benchmark swing is GBAEval: 79.6% for Claude Opus 5 and 6.7% for Gemini 3.5 Flash.
- Gemini 3.5 Flash is cheaper at $1.50 / $9 per million input/output tokens, against $5 / $25 for Claude Opus 5.
- Gemini 3.5 Flash accepts more context: 1.05M tokens versus 1M.
Side by side
| Claude Opus 5 | Gemini 3.5 Flash | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 67.8 | 54.2 |
| Released | 2026-07-24 | 2026-05-19 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 128K | 66K |
| Input $ / M tokens | $5 | $1.50 |
| Output $ / M tokens | $25 | $9 |
| Results tracked | 57 | 54 |
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 3.5 Flash: 49.4 (#49)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| DeepSWE | 73.6% | 37.4% |
| LMArena WebDev | 1691 | 1499 |
| SciCode | 56.4% | 53.1% |
| WeirdML | 91.8% | 62.6% |
| LMArena Coding | 1534 | 1492 |
| ALE-Bench | 2,165 | 911.02 |
| SWE-bench Verified | — | 79.3% |
| FrontierCode | 53.4% | — |
| CursorBench | 46.6% | — |
| FrontierSWE | 52% | — |
Agentic & Tool Use Claude Opus 5 leads
Claude Opus 5: 55.6 (#1), Gemini 3.5 Flash: 24.7 (#114)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| APEX-Agents | 65.8% | 27.5% |
| GBAEval | 79.6% | 6.7% |
| GDP.pdf | 24% | 14% |
| Vending-Bench 2 | 11,182 | 5,396 |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Banking | 48.7% | — |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
Reasoning Claude Opus 5 leads
Claude Opus 5: 77.2 (#4), Gemini 3.5 Flash: 62.8 (#18)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| ARC-AGI-2 | 90.4% | 72.1% |
| SimpleBench | 80.6% | 76.7% |
| NYT Connections (extended) | 94.3% | 92.6% |
| ARC-AGI-1 | 97.5% | 92.5% |
| CritPt | 29.1% | 13.1% |
| Chess Puzzles | 42% | 50% |
| EBR-Bench | 45.7% | 4.8% |
| LMArena Hard Prompts | 1526 | 1488 |
| Mystery Game Puzzles | 59% | 32% |
| DTBench | 97.9% | 94.7% |
| LMCA | 64.5% | 47.1% |
| Epoch Capabilities Index | 162.78 | 154.46 |
| EnigmaEval | — | 25.4% |
| Surface Evolver Bench | — | 58.1% |
| Bench to the Future 3 | 0.12 | — |
| ForecastBench | — | 59 |
Math Claude Opus 5 leads
Claude Opus 5: 86.2 (#8), Gemini 3.5 Flash: 60.7 (#36)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 85.6% | 62.8% |
| FrontierMath Tier 4 | 73.2% | 26.8% |
| OTIS Mock AIME 2024-2025 | 98.9% | 95.6% |
| ProofBench | 99% | 31% |
| LMArena Math | 1531 | 1504 |
| MathArena Final-Answer Competitions | — | 76.3% |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Too close to call
Claude Opus 5: 66.8 (#9), Gemini 3.5 Flash: 66.3 (#11)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| GPQA Diamond | 93.9% | 92.8% |
| SimpleQA Verified | 59.9% | 66.2% |
| LMArena Expert | 1557 | 1495 |
Multimodal Claude Opus 5 leads
Claude Opus 5: 50.8 (#8), Gemini 3.5 Flash: 45.7 (#15)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| LMArena Vision | 1319 | 1310 |
| Blueprint-Bench 2 | 30.4% | 33.6% |
| LMArena Document | 1516 | 1463 |
| Furniture Assembly | 60.8% | — |
Multilingual Claude Opus 5 leads
Claude Opus 5: 58.8 (#4), Gemini 3.5 Flash: 57.0 (#13)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1501 | 1476 |
| LMArena Chinese | 1574 | 1526 |
| LMArena French | 1519 | 1490 |
| LMArena German | 1524 | 1492 |
| LMArena Japanese | 1516 | 1486 |
| LMArena Korean | 1521 | 1451 |
| LMArena Russian | 1507 | 1493 |
| LMArena Spanish | 1519 | 1480 |
Instruction Following Claude Opus 5 leads
Claude Opus 5: 79.2 (#7), Gemini 3.5 Flash: 77.0 (#30)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1517 | 1467 |
Long Context Claude Opus 5 leads
Claude Opus 5: 46.5 (#21), Gemini 3.5 Flash: 45.4 (#38)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1515 | 1482 |
Writing & Preference Claude Opus 5 leads
Claude Opus 5: 79.2 (#1), Gemini 3.5 Flash: 65.5 (#47)
| Benchmark | Claude Opus 5 | Gemini 3.5 Flash |
|---|---|---|
| LMArena Text | 1507 | 1482 |
| LMArena Creative Writing | 1491 | 1470 |
| EQ-Bench 4 | 1385 | 1087 |
| LMArena Multi-Turn | 1499 | 1481 |
| EQ-Bench Creative Writing | 2133 | — |
Frequently asked questions
Is Claude Opus 5 better than Gemini 3.5 Flash?
Claude Opus 5 is the stronger model overall, scoring 67.8 to 54.2 on the Noometry Index. Gemini 3.5 Flash costs 3.0× 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 3.5 Flash?
Gemini 3.5 Flash is cheaper. It lists at $1.50 per million input tokens and $9 per million output tokens; Claude Opus 5 lists at $5 and $25.
Is Claude Opus 5 or Gemini 3.5 Flash better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 49.4 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.5 Flash does, with 1.05M tokens against 1M.
How many benchmarks do Claude Opus 5 and Gemini 3.5 Flash share?
47 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Gemini 3.5 Flash has 54.