# Claude Opus 5 vs Gemini 3.8 Flash

> Side-by-side benchmark comparison for Claude Opus 5 and Gemini 3.8 Flash across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/claude-opus-5-vs-gemini-3-8-flash
- Last updated: October 17, 2026

- Shared sourced benchmarks: 23
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Pick Claude Opus 5 if you want the stronger benchmark profile. Gemini 3.8 Flash only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Claude Opus 5 leads overall 79.63 to 69.25.
- The clearest category separation is in mathematics, where the averages are 91.6 for Claude Opus 5 and 69.7 for Gemini 3.8 Flash.
- The biggest single benchmark swing is FrontierMath v2 (Tier 4) in Mathematics, with scores of 73.171% and 21.951%.
- Gemini 3.8 Flash is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- Gemini 3.8 Flash also has the larger context window at 1M.

## Model Snapshot

| Property | Claude Opus 5 | Gemini 3.8 Flash |
|----------|----------|----------|
| Creator | Anthropic | Google |
| Type | Proprietary | Proprietary |
| Reasoning | Reasoning | Reasoning |
| Context | null | 1M |
| Overall Score | 79.63 | 69.25 |
| Benchmarks Covered | 80 | 31 |
| Pricing (input/output) | $5.00 / $25.00 | $0.75 / $3.75 |

## Category Breakdown

### Agentic

- Winner: Claude Opus 5
- Claude Opus 5 public-lane score: 78.5 (Supported · #3/128)
- Gemini 3.8 Flash public-lane score: 65.9 (Supported · #16/128)

| Benchmark | Claude Opus 5 | Gemini 3.8 Flash | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 3.0 | 42.7% | Coming soon | Coming soon |
| BrowseComp | 90.8% | Coming soon | Coming soon |
| HLE w/ tools | 64.7% | Coming soon | Coming soon |
| DeepSearchQA | 95.0% | Coming soon | Coming soon |
| DRACO | 88.6% | Coming soon | Coming soon |
| BrowseComp (10-agent, prerelease) | 93.6% | Coming soon | Coming soon |
| OSWorld 2.0 | 70.6% | 59.0% | Claude Opus 5 |
| MCP Atlas | 85.8% | Coming soon | Coming soon |
| MCP-Atlas claim coverage | 89.1% | Coming soon | Coming soon |
| LAB all-pass (Anthropic harness) | 23.58% | Coming soon | Coming soon |
| LAB criterion-pass (Anthropic harness) | 93.74% | Coming soon | Coming soon |
| LAB all-pass (Harvey held-out) | 11.7% | Coming soon | Coming soon |
| LAB criterion-pass (Harvey held-out) | 94.1% | Coming soon | Coming soon |
| Toolathlon-Verified | 80.6% | Coming soon | Coming soon |
| Toolathlon Verified Pass@3 | 87.0% | Coming soon | Coming soon |
| Toolathlon Verified Pass³ | 73.1% | Coming soon | Coming soon |
| Toolathlon Verified avg. turns | 23.5 turns | Coming soon | Coming soon |
| AutomationBench | 26.0% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 84.6% | 81.3% | Claude Opus 5 |
| ApprenticeBench | 36% | 24% | Claude Opus 5 |
| MazeBench | 12% | 4% | Claude Opus 5 |
| Finance Agent v2 | Coming soon | 61.4% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 89.4% | Coming soon |
| Terminal-Bench 4.0 | Coming soon | 19.10% | Coming soon |
| Hermes Index | Coming soon | 38.1% | Coming soon |
| Hermes Bench | Coming soon | 62.2% | Coming soon |

### Coding

- Winner: Claude Opus 5
- Claude Opus 5 public-lane score: 70.7 (Supported · #9/150)
- Gemini 3.8 Flash public-lane score: 61 (Supported · #17/150)

| Benchmark | Claude Opus 5 | Gemini 3.8 Flash | Winner |
|-----------|-----------|-----------|--------|
| Bug Hunt Bench | 27.0 fixes | 18.0 fixes | Claude Opus 5 |
| SWE-bench Verified | 96% | Coming soon | Coming soon |
| SWE-bench Pro | 79.2% | Coming soon | Coming soon |
| SWE Multilingual | 89.5% | Coming soon | Coming soon |
| SWE Multimodal | 59.4% | Coming soon | Coming soon |
| DeepSWE | 68.8% | 73.8% | Gemini 3.8 Flash |
| FrontierCode 1.1 Main | 53.4% | Coming soon | Coming soon |
| FrontierCode 1.1 Extended | 63.6% | Coming soon | Coming soon |
| FrontierSWE v2 | 52.0% | 19.6% | Claude Opus 5 |
| ProgramBench (episode 1) | 83.0% | Coming soon | Coming soon |
| ProgramBench | 93.0% | Coming soon | Coming soon |
| CursorBench 3.2 | 70.0% | 69.2% | Claude Opus 5 |
| VulcanBench v3 | 87.0% | Coming soon | Coming soon |
| VulcanBench CII v1 | 96.4% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 89.0% | 89.5% | Gemini 3.8 Flash |
| SWE-bench (Vals) | 97.0% | 80.0% | Claude Opus 5 |
| CursorBench 4.0 | 46.6% | 39.6% | Claude Opus 5 |
| PostTrainBench v1.1 | 35.0% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 89.4% | Coming soon |

### Reasoning

- Winner: Claude Opus 5
- Claude Opus 5 public-lane score: 81 (#11/28)
- Gemini 3.8 Flash public-lane score: 74.3 (#17/28)

| Benchmark | Claude Opus 5 | Gemini 3.8 Flash | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-1 | 97.50% | 98.50% | Gemini 3.8 Flash |
| ARC-AGI-2 | 90.4% | 89.2% | Claude Opus 5 |
| ARC-AGI-3 | 30.2% | 10.4% | Claude Opus 5 |

### Multimodal & Grounded

- Winner: Directional only
- Claude Opus 5 public-lane score: 92.2 (#2/54)
- Gemini 3.8 Flash public-lane score: 87.3 (#9/54)

| Benchmark | Claude Opus 5 | Gemini 3.8 Flash | Winner |
|-----------|-----------|-----------|--------|
| Chartography (no tools) | 29.6% | Coming soon | Coming soon |
| Chartography (tools) | 83.0% | Coming soon | Coming soon |
| BenchCAD Vision2Code (no tools) | 0.366 | Coming soon | Coming soon |
| BenchCAD Vision2Code (tools) | 0.821 | Coming soon | Coming soon |
| GDP.pdf (no tools) | 83.4% | Coming soon | Coming soon |
| GDP.pdf (tools) | 85.5% | Coming soon | Coming soon |
| OfficeQA | 78.1% | Coming soon | Coming soon |
| OfficeQA Pro | 66.9% | Coming soon | Coming soon |
| CharXiv w/o tools | Coming soon | 86.2% | Coming soon |
| LVBench | Coming soon | 87.1% | Coming soon |

### Knowledge

- Winner: Claude Opus 5
- Claude Opus 5 public-lane score: 82 (Supported · #7/184)
- Gemini 3.8 Flash public-lane score: 73.5 (Supported · #13/184)

| Benchmark | Claude Opus 5 | Gemini 3.8 Flash | Winner |
|-----------|-----------|-----------|--------|
| HLE | 64.7% | Coming soon | Coming soon |
| HLE w/o tools | 56.3% | Coming soon | Coming soon |
| HLE-Verified | 54.4% | 54.9% | Gemini 3.8 Flash |
| LABBench2 | 84.2% | 86.2% | Gemini 3.8 Flash |
| HealthBench (raw) | 67.1% | Coming soon | Coming soon |
| HealthBench (length-adjusted) | 57.8% | Coming soon | Coming soon |
| HealthBench Professional | 59.8% | Coming soon | Coming soon |
| HealthBench Professional (raw) | 73.4% | Coming soon | Coming soon |
| BioMysteryBench (human-solvable) | 90.1% | 88.8% | Claude Opus 5 |
| BioMysteryBench (human-difficult) | 49.4% | 56.5% | Gemini 3.8 Flash |
| SpatialBench Verified | 72.5% | Coming soon | Coming soon |
| SingleCellBench | 60.6% | Coming soon | Coming soon |
| ProteinGym Hard | 47.7% | Coming soon | Coming soon |
| Protein Design | 42.5% | Coming soon | Coming soon |
| Organic chemistry V2 | 61.6% | Coming soon | Coming soon |
| Protocols (troubleshooting) | 61.1% | Coming soon | Coming soon |
| Protocols (understanding) | 78.4% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 93.4% | 94.4% | Gemini 3.8 Flash |
| MMLU-Pro (Vals) | 91.6% | 90.2% | Claude Opus 5 |

### Multilingual

- Winner: Not comparable
- Claude Opus 5 public-lane score: Coming soon
- Gemini 3.8 Flash public-lane score: Coming soon

| Benchmark | Claude Opus 5 | Gemini 3.8 Flash | Winner |
|-----------|-----------|-----------|--------|
| GMMLU | 92.5% | Coming soon | Coming soon |
| MILU | 92.1% | Coming soon | Coming soon |
| INCLUDE | 89.8% | Coming soon | Coming soon |

### Instruction Following

- Winner: Not comparable
- Claude Opus 5 public-lane score: Coming soon
- Gemini 3.8 Flash public-lane score: Coming soon

| Benchmark | Claude Opus 5 | Gemini 3.8 Flash | Winner |
|-----------|-----------|-----------|--------|
| Gray Swan IPI (15 attempts) | 4.6% | 5.5% | Claude Opus 5 |

### Mathematics

- Winner: Not comparable
- Claude Opus 5 public-lane score: 91.6 (Unranked · 7 rankable rows)
- Gemini 3.8 Flash public-lane score: 69.7 (Unranked · 2 rankable rows)

| Benchmark | Claude Opus 5 | Gemini 3.8 Flash | Winner |
|-----------|-----------|-----------|--------|
| IMO 2026 | 42/42 | Coming soon | Coming soon |
| RiemannBench (no tools) | 60.0% | Coming soon | Coming soon |
| RiemannBench (tools) | 79.0% | Coming soon | Coming soon |
| ArXivMath Jun. 2026 (no tools) | 90.8% | Coming soon | Coming soon |
| ArXivMath Jun. 2026 (tools) | 91.3% | Coming soon | Coming soon |
| FrontierMath v2 (Tiers 1-3) | 85.614% | 68.421% | Claude Opus 5 |
| FrontierMath v2 (Tier 4) | 73.171% | 21.951% | Claude Opus 5 |

## FAQ

### Which is better overall, Claude Opus 5 or Gemini 3.8 Flash?

Claude Opus 5 is ahead overall on BenchLM right now.

### Where is the biggest gap between Claude Opus 5 and Gemini 3.8 Flash?

The widest category gap is in mathematics, where the averages are 91.6 for Claude Opus 5 and 69.7 for Gemini 3.8 Flash.

### How many benchmarks does Claude Opus 5 cover on BenchLM?

Claude Opus 5 currently has 80 sourced benchmark scores on BenchLM.

### How many benchmarks does Gemini 3.8 Flash cover on BenchLM?

Gemini 3.8 Flash currently has 31 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Claude Opus 5 vs Gemini 3.8 Flash Cyber](/compare/claude-opus-5-vs-gemini-3-8-flash-cyber)
- [Gemini 3.8 Flash vs Gemini 3.8 Flash Cyber](/compare/gemini-3-8-flash-vs-gemini-3-8-flash-cyber)
- [Claude Opus 5 vs Claude Opus 5.5](/compare/claude-opus-5-vs-claude-opus-5-5)
- [Gemini 3.8 Flash vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-gemini-3-8-flash)

## Explore More

- [Claude Opus 5 profile](/models/claude-opus-5)
- [Gemini 3.8 Flash profile](/models/gemini-3-8-flash)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)