Model comparison
Claude Opus 4.8 vs Claude Opus 5
Head-to-head evidence from 32 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Claude Opus 4.8 #6 (Supported); Claude Opus 5 #1 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Claude Opus 4.8 and Claude Opus 5 share 32 comparable benchmark results. 5 of 8 categories are comparable. 23 results are unique to Claude Opus 4.8; 47 to Claude Opus 5.
Updated July 24, 2026- Shared results
- 32
- Claude Opus 4.8 only
- 23
- Claude Opus 5 only
- 47
- Comparable categories
- 5 / 8
Pick Claude Opus 5 if you want the stronger benchmark profile. Claude Opus 4.8 only becomes the better choice if multimodal & grounded is the priority.
Confidence note. This is a partial-evidence comparison with 32 shared benchmark results across 6 evidence categories; 5 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Claude Opus 5 is clearly ahead on the BenchAlign aggregate, 85.88 to 77.44. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Claude Opus 5's sharpest advantage is in reasoning, where it averages 90.4 against 72.1. The single biggest benchmark swing on the page is ARC-AGI-2, 72.1% to 90.4%. Claude Opus 4.8 does hit back in multimodal & grounded, so the answer changes if that is the part of the workload you care about most.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Claude Opus 4.8 | Δ | Claude Opus 5 |
|---|---|---|---|
| Reasoning | Claude Opus 4.872.1 | Margin→ 18.3 | Claude Opus 590.4 |
| Agentic | Claude Opus 4.880.3 | Margin→ 10.5 | Claude Opus 590.8 |
| Multimodal | Claude Opus 4.877.0 | Margin← 10.1 | Claude Opus 566.9 |
| Coding | Claude Opus 4.881.1 | Margin→ 8.4 | Claude Opus 589.5 |
| Knowledge | Claude Opus 4.862.7 | Margin→ 2.0 | Claude Opus 564.7 |
| Math | Claude Opus 4.853.9 | MarginNo overlap | Claude Opus 5Not measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
ARC-AGI-2
ReasoningA 72.1%B 90.4%Winner: Claude Opus 5Δ 18.3ARC-AGI-2: Claude Opus 4.8 scored 72.1%; Claude Opus 5 scored 90.4%. Claude Opus 5 wins this benchmark. - Source ↗
SWE-bench Pro
CodingA 69.2%B 79.2%Winner: Claude Opus 5Δ 10SWE-bench Pro: Claude Opus 4.8 scored 69.2%; Claude Opus 5 scored 79.2%. Claude Opus 5 wins this benchmark. - Source ↗
SWE-bench Verified
CodingA 88.6%B 96%Winner: Claude Opus 5Δ 7.4SWE-bench Verified: Claude Opus 4.8 scored 88.6%; Claude Opus 5 scored 96%. Claude Opus 5 wins this benchmark. - Source ↗
HLE
KnowledgeA 57.9%B 64.7%Winner: Claude Opus 5Δ 6.8HLE: Claude Opus 4.8 scored 57.9%; Claude Opus 5 scored 64.7%. Claude Opus 5 wins this benchmark. - Source ↗
BrowseComp
AgenticA 84.3%B 90.8%Winner: Claude Opus 5Δ 6.5BrowseComp: Claude Opus 4.8 scored 84.3%; Claude Opus 5 scored 90.8%. Claude Opus 5 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Claude Opus 4.8 | Claude Opus 5 | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Claude Opus 4.8$5 input / $25 output | Claude Opus 5$5 input / $25 output | Listed prices are equal. |
| Generation speedtokens per second | Claude Opus 4.8Not available | Claude Opus 5Not available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Claude Opus 4.8Not available | Claude Opus 5Not available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Claude Opus 4.81M | Claude Opus 5Not available | A complete context comparison is not available. |
Benchmark Deep Dive
AgenticClaude Opus 5 wins35 benchmarks
| Benchmark | Claude Opus 4.8 | Claude Opus 5 | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 74.6% | — | Not comparable |
| BrowseCompSource | 84.3% | 90.8% | Claude Opus 5 leads |
| DeepSearchQASource | 93.1% | 95.0% | Claude Opus 5 leads |
| OSWorld-VerifiedSource | 83.4% | — | Not comparable |
| Finance Agent v2Source | 53.9% | — | Not comparable |
| GDPval-AASource | 1593 | 1861 | Claude Opus 5 leads |
| MCP AtlasSource | 82.2% | 85.8% | Claude Opus 5 leads |
| ToolathlonSource | 59.9% | — | Not comparable |
| Gert LabsSource | 72.97% | — | Not comparable |
| AA Agentic IndexSource | 47.2% | 55.3% | Claude Opus 5 leads |
| τ²-bench resultsSource | 94.4% | — | Not comparable |
| GDPval-AASource | 54.6% | 68.0% | Claude Opus 5 leads |
| ResearchClawBenchSource | 21.1% | — | Not comparable |
| OSWorld 2.0Source | 20.6% | 70.6% | Claude Opus 5 leads |
| AA BriefcaseSource | 1346 | 1720 | Claude Opus 5 leads |
| AA AutomationBenchSource | 48.5% | — | Not comparable |
| AA EnterpriseOps-GymSource | 44.0% | — | Not comparable |
| AA Harvey LABSource | 91.1% | — | Not comparable |
| AA Tau3 BankingSource | 27.6% | 30.3% | Claude Opus 5 leads |
| terminalBenchHardSource | 58.3% | — | Not comparable |
| aaTerminalBench21Source | 84.6% | 89.1% | Claude Opus 5 leads |
| FrontierBench v0.1Source | — | 43.3% | Not comparable |
| HLE w/ toolsSource | — | 64.7% | Not comparable |
| DRACOSource | — | 88.6% | Not comparable |
| BrowseComp (10-agent, prerelease)Source | — | 93.6% | Not comparable |
| MCP-Atlas claim coverageSource | — | 89.1% | Not comparable |
| LAB all-pass (Anthropic harness)Source | — | 23.58% | Not comparable |
| LAB criterion-pass (Anthropic harness)Source | — | 93.74% | Not comparable |
| LAB all-pass (Harvey held-out)Source | — | 11.7% | Not comparable |
| LAB criterion-pass (Harvey held-out)Source | — | 94.1% | Not comparable |
| Toolathlon-VerifiedSource | — | 80.6% | Not comparable |
| Toolathlon Verified Pass@3Source | — | 87.0% | Not comparable |
| Toolathlon Verified Pass³Source | — | 73.1% | Not comparable |
| Toolathlon Verified avg. turnsSource | — | 23.5 turns | Not comparable |
| AutomationBenchSource | — | 26.0% | Not comparable |
CodingClaude Opus 5 wins14 benchmarks
| Benchmark | Claude Opus 4.8 | Claude Opus 5 | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 88.6% | 96% | Claude Opus 5 leads |
| SWE-bench ProSource | 69.2% | 79.2% | Claude Opus 5 leads |
| SWE MultilingualSource | 84.4% | 89.5% | Claude Opus 5 leads |
| SWE MultimodalSource | 38.4% | 59.4% | Claude Opus 5 leads |
| Terminal-Bench 2.0Source | 74.6% | — | Not comparable |
| cursorBench31Source | 58.4% | — | Not comparable |
| cursorBench32Source | 62.3% | 70.0% | Claude Opus 5 leads |
| AA Coding IndexSource | 74.3% | 78.0% | Claude Opus 5 leads |
| AA-SciCodeSource | 53.5% | 55.7% | Claude Opus 5 leads |
| FrontierCode 1.1 MainSource | 46.5% | 53.4% | Claude Opus 5 leads |
| deepSweSource | — | 68.8% | Not comparable |
| FrontierCode 1.1 ExtendedSource | — | 63.6% | Not comparable |
| ProgramBench (episode 1)Source | — | 83.0% | Not comparable |
| ProgramBenchSource | — | 93.0% | Not comparable |
ReasoningClaude Opus 5 wins5 benchmarks
KnowledgeClaude Opus 5 wins23 benchmarks
| Benchmark | Claude Opus 4.8 | Claude Opus 5 | Result |
|---|---|---|---|
| GPQASource | 93.6% | — | Not comparable |
| GPQA-DSource | 93.6% | — | Not comparable |
| HLESource | 57.9% | 64.7% | Claude Opus 5 leads |
| HLE w/o toolsSource | 49.8% | 56.3% | Claude Opus 5 leads |
| Artificial Analysis Intelligence IndexSource | 55.7% | 60.7% | Claude Opus 5 leads |
| AA-GPQA DiamondSource | 92.0% | 93.2% | Claude Opus 5 leads |
| AA-HLESource | 45.7% | 52.6% | Claude Opus 5 leads |
| AA-Omniscience IndexSource | 27.4% | 31.3% | Claude Opus 5 leads |
| AA-Omniscience AccuracySource | 46.6% | 54.2% | Claude Opus 5 leads |
| AA-Omniscience Hallucination RateSource | 35.9% | 50.1% | Claude Opus 4.8 leads |
| HealthBench (raw)Source | — | 67.1% | Not comparable |
| HealthBench (length-adjusted)Source | — | 57.8% | Not comparable |
| HealthBench ProfessionalSource | — | 59.8% | Not comparable |
| HealthBench Professional (raw)Source | — | 73.4% | Not comparable |
| BioMysteryBench (human-solvable)Source | — | 90.1% | Not comparable |
| BioMysteryBench (human-difficult)Source | — | 49.4% | Not comparable |
| SpatialBench VerifiedSource | — | 72.5% | Not comparable |
| SingleCellBenchSource | — | 60.6% | Not comparable |
| ProteinGym HardSource | — | 47.7% | Not comparable |
| Protein DesignSource | — | 42.5% | Not comparable |
| Organic chemistry V2Source | — | 61.6% | Not comparable |
| Protocols (troubleshooting)Source | — | 61.1% | Not comparable |
| Protocols (understanding)Source | — | 78.4% | Not comparable |
Math8 benchmarks
| Benchmark | Claude Opus 4.8 | Claude Opus 5 | Result |
|---|---|---|---|
| USAMO 2026Source | 96.7% | — | Not comparable |
| FrontierMath v2 (Tiers 1-3)Source | 47.241% | — | Not comparable |
| FrontierMath v2 (Tier 4)Source | 31.250% | — | Not comparable |
| IMO 2026Source | — | 42/42 | Not comparable |
| RiemannBench (no tools)Source | — | 60.0% | Not comparable |
| RiemannBench (tools)Source | — | 79.0% | Not comparable |
| ArXivMath Jun. 2026 (no tools)Source | — | 90.8% | Not comparable |
| ArXivMath Jun. 2026 (tools)Source | — | 91.3% | Not comparable |
Multilingual3 benchmarks
MultimodalClaude Opus 4.8 wins13 benchmarks
| Benchmark | Claude Opus 4.8 | Claude Opus 5 | Result |
|---|---|---|---|
| OfficeQA ProSource | 66.2% | 66.9% | Claude Opus 5 leads |
| ScreenSpot ProSource | 87.9% | — | Not comparable |
| CharXivSource | 89.9% | — | Not comparable |
| CharXiv w/o toolsSource | 80.5% | — | Not comparable |
| Design Arena WebsiteSource | 1270 | — | Not comparable |
| Chartography (no tools)Source | — | 29.6% | Not comparable |
| Chartography (tools)Source | — | 83.0% | Not comparable |
| BenchCAD Vision2Code (no tools)Source | — | 0.366 | Not comparable |
| BenchCAD Vision2Code (tools)Source | — | 0.821 | Not comparable |
| GDP.pdf (no tools)Source | — | 83.4% | Not comparable |
| GDP.pdf (tools)Source | — | 85.5% | Not comparable |
| OfficeQASource | — | 78.1% | Not comparable |
| AA-MMMU-ProSource | — | 84.7% | Not comparable |
Inst. Following1 benchmarks
| Benchmark | Claude Opus 4.8 | Claude Opus 5 | Result |
|---|---|---|---|
| AA-IFBenchSource | 62.2% | — | Not comparable |
Frequently Asked Questions (6)
Which is better, Claude Opus 4.8 or Claude Opus 5?
Claude Opus 5 is ahead on BenchLM's BenchAlign leaderboard, 85.88 to 77.44. The biggest single separator in this matchup is ARC-AGI-2, where the scores are 72.1% and 90.4%.
Which is better for knowledge tasks, Claude Opus 4.8 or Claude Opus 5?
Claude Opus 5 has the edge for knowledge tasks in this comparison, averaging 64.7 versus 62.7. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.
Which is better for coding, Claude Opus 4.8 or Claude Opus 5?
Claude Opus 5 has the edge for coding in this comparison, averaging 89.5 versus 81.1. Inside this category, SWE Multimodal is the benchmark that creates the most daylight between them.
Which is better for reasoning, Claude Opus 4.8 or Claude Opus 5?
Claude Opus 5 has the edge for reasoning in this comparison, averaging 90.4 versus 72.1. Inside this category, ARC-AGI-3 is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, Claude Opus 4.8 or Claude Opus 5?
Claude Opus 5 has the edge for agentic tasks in this comparison, averaging 90.8 versus 80.3. Inside this category, AA Briefcase is the benchmark that creates the most daylight between them.
Which is better for multimodal and grounded tasks, Claude Opus 4.8 or Claude Opus 5?
Claude Opus 4.8 has the edge for multimodal and grounded tasks in this comparison, averaging 77 versus 66.9. Inside this category, OfficeQA Pro is the benchmark that creates the most daylight between them.