Claude Opus 5 vs Inkling 2 Small
New ComparisonScores are fetched live from the current database on every visit — nothing is cached.
Winner
Claude Opus 5 wins this comparison with an overall score of 88 vs 60 for Inkling 2 Small, leading 5 of 7 shared benchmarks.
Based on 7 of 85 benchmarks scored by all 2 selected models.
Benchmark Scores Comparison
Claude Opus 5
Score: 40.0
Inkling 2 Small
594.1B params
Coverage matrix
2 models × 85 benchmarks — hover a benchmark name for its category
| Benchmark | Inkling 2 Small Thinking Machines | Claude Opus 5 Anthropic |
|---|---|---|
| OSWorld-G | — | 76.8 |
| GPQA Diamond | 93 | 93.7 |
| ArXivMath | — | 71.5 |
| Hy-LifeSearch (Internal) | — | 56.1 |
| Automation-Bench | — | 48.7 |
| SWE-bench Pro | — | 79.2 |
| BFCL-V4 | — | 80.6 |
| BioMysteryBench | 70 | 72.1 |
| Agents' Last Exam | — | 25.1 |
| Toolathlon Verified | — | 76.5 |
| Hy-Backend 2.0 (Internal) | — | 40.3 |
| TAU3-Bench | 40.4 | — |
| ScreenSpot Pro | 83.8 | — |
| Hy-SWE Max Verified (Internal) | — | 70.1 |
| Hy-BrowseComp-Pro2 (Internal) | — | 61.3 |
| MuirBench | 79.2 | — |
| GDPVal | — | 80 |
| OfficeQA Pro | — | 66.9 |
| DeepSearch QA | 90.9 | — |
| SWE Atlas - TW | — | 68.7 |
| MLE-bench Lite | — | 88 |
| LOCA-Bench (256k) | — | 53.7 |
| SWE Atlas - QnA | — | 57.5 |
| ApexBench | — | 65.8 |
| NL2Repo | — | 75.3 |
| MLS-Bench | — | 49.8 |
| MMMU-Pro | 81.6 | — |
| Chartography | — | 84 |
| PostTrainBench | — | 35 |
| BabyVision | — | 94.1 |
| OSWorld V2.1 | — | 74 |
| Apex-Agents | — | 41.8 |
| WorkSpaceBench | — | 75 |
| Terminal-Bench Science 0.1 | 5.7 | — |
| BankerToolBench | — | 81.9 |
| E-Bench-Code (Internal) | — | 82.7 |
| OSWorld-Verified | — | 86.4 |
| DeepresearchBII | — | 54.1 |
| Workspace-Bench-Lite | — | 68.1 |
| SpreadsheetBench 2 | — | 33.6 |
| HLE (with tools) | — | 60.9 |
| CharXiv (RQ) | 85.6 | 83.7 |
| GDP.pdf | — | 21.6 |
| CritPt (no tools) | — | 29.1 |
| BrowseComp | — | 90.8 |
| NL2Repo-Bench | — | 75.3 |
| DRACO | — | 88.6 |
| SkillsBench | — | 63.7 |
| WebArena-Verified | — | 69.1 |
| SWE-bench Multilingual | 81.2 | 89.5 |
| Harvey Legal Agent | — | 22.7 |
| PaperBench | — | 86.9 |
| Agents' Last Exam (CLI) | — | 29.5 |
| Hy-FinAgentBench (Internal) | — | 82 |
| Humanity's Last Exam | — | 54.9 |
| Global-MMLU-Lite | 91 | — |
| SUPERChem | — | 76.7 |
| HorizonMath (pass@4) | — | 5.3 |
| Terminal-Bench 2.1 (Terminus-2) | 81.3 | 86.7 |
| DeepSWE | 72.1 | 68.8 |
| BrokenArXiv | — | 77.7 |
| HMMT Feb 26 | 95.8 | — |
| Hy-FinmodelBench v2 (Internal) | — | 66 |
| MathArena Apex 2025 | — | 91.4 |
| JobBench | — | 68 |
| $OneMillion-Bench | — | 68.1 |
| OSWorld 2.0 (Binary) | — | 70.6 |
| WideSearch | — | 84 |
| SimpleVQA | 68.2 | — |
| SkillsBench Avg5 | — | 63.7 |
| Vision2Web | — | 80.2 |
| TauBench V3 Banking | — | 48.7 |
| E-Bench (Internal) | — | 77.8 |
| ProgramBench | — | 39.5 |
| MCP-Atlas | 85 | 85.7 |
| SWE-Marathon | — | 50 |
| Terminal-Bench 4.0 | — | 51.8 |
| SWE Atlas - RF | — | 60 |
| DeepSWE 1.1 | — | 74 |
| Hy-CompanyBench V2 (Internal) | — | 72.7 |
| Harbor-Index | — | 56.9 |
| SWE Refactor Bench | — | 31 |
| GDPVal-AA v2 | — | 1831 |
| WeaveBench | — | 86.4 |
| ZEROBench | — | 52 |
| Covered | 16/85 | 76/85 |
Scorecard
Overall 0–100 score — category strengths (50%), shared-benchmark wins (30%), coverage (20%)
| Model | Overall score | Shared benchmark wins | Coverage | Strongest at | Weakest at |
|---|---|---|---|---|---|
| Claude Opus 5 Winner | 88.3 | 5 of 7 | 76 of 85 (89.0%) | STEM & Reasoning | Document Understanding |
| Inkling 2 Small | 60 | 2 of 7 | 16 of 85 (19.0%) | Document Understanding | STEM & Reasoning |
Category strengths
Where each model stands, per benchmark category — score relative to the best model, adjusted for benchmark coverage. This also ranks models when they share few benchmarks.
| Category | Claude Opus 5 | Inkling 2 Small |
|---|---|---|
| General Agents 34 benchmarks | 100 | 99 |
| Agentic Coding 24 benchmarks | 98.5 | 94 |
| STEM & Reasoning 11 benchmarks | 100 | 93.2 |
| Document Understanding 1 benchmark | 90.3 | 100 |
Fully comparable benchmarks
Scored by all 2 selected models (7)
| Benchmark | Category | Claude Opus 5 | Inkling 2 Small |
|---|---|---|---|
| GPQA Diamond | stem_reasoning | 93.7 | 93 |
| BioMysteryBench | stem_reasoning | 72.1 | 70 |
| CharXiv (RQ) | document_understanding | 83.7 | 85.6 |
| SWE-bench Multilingual | coding_agent | 89.5 | 81.2 |
| Terminal-Bench 2.1 (Terminus-2) | coding_agent | 86.7 | 81.3 |
| DeepSWE | coding_agent | 68.8 | 72.1 |
| MCP-Atlas | general_agent | 85.7 | 85 |
Partial coverage benchmarks
Scored by at least one but not all selected models (78) — missing scores show as —
| Benchmark | Category | Claude Opus 5 | Inkling 2 Small |
|---|---|---|---|
| OSWorld-G 1/2 models | general_agent | 76.8 | — |
| ArXivMath 1/2 models | stem_reasoning | 71.5 | — |
| Hy-LifeSearch (Internal) 1/2 models | general_agent | 56.1 | — |
| Automation-Bench 1/2 models | general_agent | 48.7 | — |
| SWE-bench Pro 1/2 models | coding_agent | 79.2 | — |
| BFCL-V4 1/2 models | general_agent | 80.6 | — |
| Agents' Last Exam 1/2 models | general_agent | 25.1 | — |
| Toolathlon Verified 1/2 models | general_agent | 76.5 | — |
| Hy-Backend 2.0 (Internal) 1/2 models | coding_agent | 40.3 | — |
| TAU3-Bench 1/2 models | general_agent | — | 40.4 |
| ScreenSpot Pro 1/2 models | general_agent | — | 83.8 |
| Hy-SWE Max Verified (Internal) 1/2 models | coding_agent | 70.1 | — |
| Hy-BrowseComp-Pro2 (Internal) 1/2 models | general_agent | 61.3 | — |
| MuirBench 1/2 models | video_understanding | — | 79.2 |
| GDPVal 1/2 models | general_agent | 80 | — |
| OfficeQA Pro 1/2 models | general_agent | 66.9 | — |
| DeepSearch QA 1/2 models | general_agent | — | 90.9 |
| SWE Atlas - TW 1/2 models | coding_agent | 68.7 | — |
| MLE-bench Lite 1/2 models | coding_agent | 88 | — |
| LOCA-Bench (256k) 1/2 models | long_context | 53.7 | — |
| SWE Atlas - QnA 1/2 models | coding_agent | 57.5 | — |
| ApexBench 1/2 models | multimodal_agent | 65.8 | — |
| NL2Repo 1/2 models | coding_agent | 75.3 | — |
| MLS-Bench 1/2 models | coding_agent | 49.8 | — |
| MMMU-Pro 1/2 models | vision_language | — | 81.6 |
| Chartography 1/2 models | multimodal_agent | 84 | — |
| PostTrainBench 1/2 models | coding_agent | 35 | — |
| BabyVision 1/2 models | vision_language | 94.1 | — |
| OSWorld V2.1 1/2 models | general_agent | 74 | — |
| Apex-Agents 1/2 models | general_agent | 41.8 | — |
| WorkSpaceBench 1/2 models | general_agent | 75 | — |
| Terminal-Bench Science 0.1 1/2 models | coding_agent | — | 5.7 |
| BankerToolBench 1/2 models | general_agent | 81.9 | — |
| E-Bench-Code (Internal) 1/2 models | coding_agent | 82.7 | — |
| OSWorld-Verified 1/2 models | general_agent | 86.4 | — |
| DeepresearchBII 1/2 models | general_agent | 54.1 | — |
| Workspace-Bench-Lite 1/2 models | general_agent | 68.1 | — |
| SpreadsheetBench 2 1/2 models | general_agent | 33.6 | — |
| HLE (with tools) 1/2 models | stem_reasoning | 60.9 | — |
| GDP.pdf 1/2 models | long_context | 21.6 | — |
| CritPt (no tools) 1/2 models | stem_reasoning | 29.1 | — |
| BrowseComp 1/2 models | general_agent | 90.8 | — |
| NL2Repo-Bench 1/2 models | coding_agent | 75.3 | — |
| DRACO 1/2 models | general_agent | 88.6 | — |
| SkillsBench 1/2 models | general_agent | 63.7 | — |
| WebArena-Verified 1/2 models | general_agent | 69.1 | — |
| Harvey Legal Agent 1/2 models | general_agent | 22.7 | — |
| PaperBench 1/2 models | coding_agent | 86.9 | — |
| Agents' Last Exam (CLI) 1/2 models | general_agent | 29.5 | — |
| Hy-FinAgentBench (Internal) 1/2 models | domain_finance | 82 | — |
| Humanity's Last Exam 1/2 models | stem_reasoning | 54.9 | — |
| Global-MMLU-Lite 1/2 models | multilingual | — | 91 |
| SUPERChem 1/2 models | stem_reasoning | 76.7 | — |
| HorizonMath (pass@4) 1/2 models | stem_reasoning | 5.3 | — |
| BrokenArXiv 1/2 models | stem_reasoning | 77.7 | — |
| HMMT Feb 26 1/2 models | stem_reasoning | — | 95.8 |
| Hy-FinmodelBench v2 (Internal) 1/2 models | domain_finance | 66 | — |
| MathArena Apex 2025 1/2 models | stem_reasoning | 91.4 | — |
| JobBench 1/2 models | general_agent | 68 | — |
| $OneMillion-Bench 1/2 models | general_capabilities | 68.1 | — |
| OSWorld 2.0 (Binary) 1/2 models | general_agent | 70.6 | — |
| WideSearch 1/2 models | general_agent | 84 | — |
| SimpleVQA 1/2 models | vision_language | — | 68.2 |
| SkillsBench Avg5 1/2 models | coding_agent | 63.7 | — |
| Vision2Web 1/2 models | vision_language | 80.2 | — |
| TauBench V3 Banking 1/2 models | general_agent | 48.7 | — |
| E-Bench (Internal) 1/2 models | general_agent | 77.8 | — |
| ProgramBench 1/2 models | coding_agent | 39.5 | — |
| SWE-Marathon 1/2 models | coding_agent | 50 | — |
| Terminal-Bench 4.0 1/2 models | coding_agent | 51.8 | — |
| SWE Atlas - RF 1/2 models | coding_agent | 60 | — |
| DeepSWE 1.1 1/2 models | coding_agent | 74 | — |
| Hy-CompanyBench V2 (Internal) 1/2 models | general_agent | 72.7 | — |
| Harbor-Index 1/2 models | coding_agent | 56.9 | — |
| SWE Refactor Bench 1/2 models | coding_agent | 31 | — |
| GDPVal-AA v2 1/2 models | general_agent | 1831 | — |
| WeaveBench 1/2 models | vision_language | 86.4 | — |
| ZEROBench 1/2 models | vision_language | 52 | — |
Comparison FAQ
Which model is better, Claude Opus 5 or Inkling 2 Small?
Claude Opus 5 wins this comparison with an overall score of 88 vs 60 for Inkling 2 Small, leading 5 of 7 shared benchmarks.
Which model is better at Agentic Coding, Claude Opus 5 or Inkling 2 Small?
Claude Opus 5 leads the Agentic Coding category (strength score 98.5/100 across 3 comparable benchmarks).
Where does Inkling 2 Small outperform Claude Opus 5?
Inkling 2 Small leads the Document Understanding category (strength score 100/100 across 1 comparable benchmark).
How reliable is this comparison verdict?
Based on 7 of 85 benchmarks scored by all 2 selected models.