Model comparison
Claude Opus 4 vs Gemini 3.8 Flash
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 43.1 on the Noometry Index.
Last verified . 29 shared benchmarks.
GoogleSummary
- They share 29 benchmarks with published results for both. Claude Opus 4 scores higher in 0 categories and Gemini 3.8 Flash in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 27.3.
- The biggest single-benchmark swing is ARC-AGI-2: 8.6% for Claude Opus 4 and 89.2% for Gemini 3.8 Flash.
- Gemini 3.8 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 200K.
Side by side
| Claude Opus 4 | Gemini 3.8 Flash | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 43.1 | 61.8 |
| Released | 2025-05-22 | 2026-09-02 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1.05M |
| Max output | 32K | 66K |
| Input $ / M tokens | $15 | $0.75 |
| Output $ / M tokens | $75 | $3.75 |
| Results tracked | 56 | 50 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.8 Flash leads
Claude Opus 4: 47.2 (#62), Gemini 3.8 Flash: 59.2 (#15)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| WeirdML | 43.7% | 84.8% |
| LMArena Coding | 1442 | 1510 |
| SWE-bench Verified | 70.7% | — |
| DeepSWE | — | 73.8% |
| FrontierCode | — | 41.2% |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| CursorBench | — | 39.6% |
| LMArena WebDev | — | 1584 |
| FrontierSWE | — | 19.6% |
| SciCode | — | 56.6% |
| GSO | 6.9% | — |
| ALE-Bench | — | 1,270 |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Gemini 3.8 Flash leads
Claude Opus 4: 34.8 (#42), Gemini 3.8 Flash: 41.8 (#21)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| APEX-Agents | — | 64.3% |
| Remote Labor Index | — | 5.8% |
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| GDP.pdf | — | 23.4% |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |
| Vending-Bench 2 | — | 5,094 |
Reasoning Gemini 3.8 Flash leads
Claude Opus 4: 27.3 (#121), Gemini 3.8 Flash: 76.9 (#5)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| ARC-AGI-2 | 8.6% | 89.2% |
| ARC-AGI-1 | 35.7% | 98.5% |
| CritPt | 0.3% | 18.3% |
| LMArena Hard Prompts | 1399 | 1508 |
| DTBench | 81.6% | 95.7% |
| LMCA | 37.4% | 52.9% |
| Epoch Capabilities Index | 142.67 | 156.71 |
| SimpleBench | 58.8% | — |
| Kagi LLM Benchmark | 74.3% | — |
| NYT Connections (extended) | — | 97.4% |
| Chess Puzzles | — | 61% |
| EnigmaEval | 5.6% | — |
| Mystery Game Puzzles | — | 47% |
| Surface Evolver Bench | — | 76.9% |
| ForecastBench | 61.1 | — |
Math Gemini 3.8 Flash leads
Claude Opus 4: 42.0 (#86), Gemini 3.8 Flash: 65.3 (#28)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 98.9% |
| LMArena Math | 1390 | 1528 |
| FrontierMath (Tiers 1-3) | — | 68.4% |
| FrontierMath Tier 4 | — | 22% |
| ProofBench | — | 48% |
| Omni-MATH | 61.6% | — |
| MATH Level 5 | 85% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Gemini 3.8 Flash leads
Claude Opus 4: 44.0 (#88), Gemini 3.8 Flash: 74.8 (#2)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| GPQA Diamond | 76.3% | 95.4% |
| Humanity's Last Exam | 10.7% | 44.5% |
| LMArena Expert | 1386 | 1524 |
| SimpleQA Verified | — | 69.7% |
| MMLU-Pro | 87.5% | — |
| Confabulations | 15.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 70.8% | — |
Multimodal Gemini 3.8 Flash leads
Claude Opus 4: 31.5 (#106), Gemini 3.8 Flash: 40.7 (#45)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Vision | 1192 | 1314 |
| GeoBench | 49% | — |
| VPCT | 38% | — |
| Blueprint-Bench 2 | — | 38.6% |
| Furniture Assembly | — | 31.7% |
Multilingual Gemini 3.8 Flash leads
Claude Opus 4: 48.8 (#138), Gemini 3.8 Flash: 58.0 (#5)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Non-English | 1362 | 1491 |
| LMArena Chinese | 1386 | 1554 |
| LMArena French | 1372 | 1498 |
| LMArena German | 1391 | 1493 |
| LMArena Japanese | 1331 | 1502 |
| LMArena Korean | 1321 | 1459 |
| LMArena Russian | 1392 | 1515 |
| LMArena Spanish | 1389 | 1485 |
Instruction Following Too close to call
Claude Opus 4: 77.1 (#28), Gemini 3.8 Flash: 78.0 (#13)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Instruction Following | 1406 | 1490 |
| IFEval | 91.8% | — |
Long Context Gemini 3.8 Flash leads
Claude Opus 4: 39.6 (#172), Gemini 3.8 Flash: 46.3 (#24)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Longer Query | 1422 | 1508 |
| Fiction.LiveBench | 61.1% | — |
Writing & Preference Gemini 3.8 Flash leads
Claude Opus 4: 61.2 (#89), Gemini 3.8 Flash: 72.2 (#15)
| Benchmark | Claude Opus 4 | Gemini 3.8 Flash |
|---|---|---|
| LMArena Text | 1377 | 1499 |
| LMArena Creative Writing | 1387 | 1492 |
| EQ-Bench Creative Writing | 1580 | 1748 |
| LMArena Multi-Turn | 1396 | 1501 |
| Short-Story Creative Writing | 83.6% | — |
| WildBench | 85.2% | — |
Frequently asked questions
Is Claude Opus 4 better than Gemini 3.8 Flash?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 43.1 on the Noometry Index.
Which is cheaper, Claude Opus 4 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 4 lists at $15 and $75.
Is Claude Opus 4 or Gemini 3.8 Flash better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 47.2 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.8 Flash does, with 1.05M tokens against 200K.
How many benchmarks do Claude Opus 4 and Gemini 3.8 Flash share?
29 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and Gemini 3.8 Flash has 50.