DeepSeek V4 Pro vs Claude Opus 5 vs MiniMax-M3.1-Flash vs Hy4 flash vs Ripple-65B Preview
New ComparisonScores are fetched live from the current database on every visit — nothing is cached.
Winner
Claude Opus 5 scores higher overall across category strength profiles (86 vs 58 for DeepSeek V4 Pro); only 0 benchmarks are directly comparable.
Only 0 of 125 benchmarks are scored by every selected model — the verdict leans on category strength profiles and is indicative.
Benchmark Scores Comparison
DeepSeek V4 Pro
Score: 15.0
Claude Opus 5
Score: 40.0
MiniMax-M3.1-Flash
428.0B params
Hy4 flash
220.0B params
Ripple-65B Preview
65.0B params
Coverage matrix
5 models × 125 benchmarks — hover a benchmark name for its category
| Benchmark | MiniMax-M3.1-Flash MiniMaxAI | Ripple-65B Preview Ricursive Intelligence | Hy4 flash Tencent | DeepSeek V4 Pro DeepSeek | Claude Opus 5 Anthropic |
|---|---|---|---|---|---|
| AGIEval (Chinese) | — | — | 90.6 | — | — |
| SVG-Bench | — | — | — | 61.7 | — |
| OSWorld-G | — | — | — | — | 76.8 |
| GPQA Diamond | — | 85.2 | — | 92.8 | 93.7 |
| CountBenchQA | — | — | 86.6 | — | — |
| OCW Courses | — | — | 64 | — | — |
| ArXivMath | — | — | — | 62.1 | 71.5 |
| Chinese-SimpleQA | — | — | 74.6 | — | — |
| Hy-LifeSearch (Internal) | — | — | — | 46.9 | 56.1 |
| Automation-Bench | 62.8 | — | — | 30.4 | 48.7 |
| SWE-bench Pro | — | 61.42 | — | 55.4 | 79.2 |
| BFCL-V4 | — | — | — | — | 80.6 |
| RealWorldQA | — | — | 71.1 | — | — |
| BioMysteryBench | — | — | — | 61.6 | 72.1 |
| AA-LCR | — | 75 | — | — | — |
| AIME 26 | — | 95 | — | — | — |
| Agents' Last Exam | 29.6 | — | — | 21.9 | 25.1 |
| Toolathlon Verified | — | — | — | 74.1 | 76.5 |
| Hy-Backend 2.0 (Internal) | — | — | — | 33.8 | 40.3 |
| Hy-SWE Max Verified (Internal) | — | — | — | 65.7 | 70.1 |
| Hy-BrowseComp-Pro2 (Internal) | — | — | — | 46.5 | 61.3 |
| ARC-c | — | — | 95 | — | — |
| AutomationBench (Public) | — | 36.4 | — | — | — |
| GDPVal | — | — | — | 70.3 | 80 |
| OfficeQA Pro | — | — | — | 65.4 | 66.9 |
| PIQA | — | — | 84.4 | — | — |
| StrongREJECT v2 (adaptive) | — | 97.5 | — | — | — |
| SWE Atlas - TW | — | — | — | 45.6 | 68.7 |
| MLE-bench Lite | — | — | — | — | 88 |
| MBPP+ | — | — | 73.7 | — | — |
| LOCA-Bench (256k) | 60.3 | — | — | — | 53.7 |
| WorkBuddy Blind User Evaluation (satisfaction) | — | — | 2.79 | — | — |
| SWE-bench Pro v2.0 Hard | — | 79.1 | — | — | — |
| SWE Atlas - QnA | — | — | — | 53.4 | 57.5 |
| ApexBench | — | — | — | — | 65.8 |
| TextVQA | — | — | 73.4 | — | — |
| NL2Repo | — | — | — | 61.5 | 75.3 |
| MLS-Bench | — | — | — | — | 49.8 |
| MMMU-Pro | — | — | 52 | — | — |
| Chartography | — | — | — | — | 84 |
| PostTrainBench | 39.3 | — | — | 24.5 | 35 |
| INCLUDE | — | — | 82.8 | — | — |
| BabyVision | 91 | — | — | — | 94.1 |
| Claw-Eval Avg | — | — | — | 58.4 | — |
| OSWorld V2.1 | — | — | — | — | 74 |
| Apex-Agents | — | — | — | 32.4 | 41.8 |
| WorkSpaceBench | — | — | — | 65.4 | 75 |
| MMLU-Pro | — | 84.2 | 71.1 | — | — |
| Terminal-Bench Science 0.1 | 3.3 | — | — | — | — |
| BankerToolBench | 76.2 | — | — | 73.1 | 81.9 |
| SuperGPQA | — | — | 57.8 | — | — |
| E-Bench-Code (Internal) | — | — | — | 64.3 | 82.7 |
| OSWorld-Verified | — | — | — | — | 86.4 |
| CV-Bench | — | — | 83.4 | — | — |
| DeepresearchBII | — | — | — | — | 54.1 |
| Workspace-Bench-Lite | — | — | — | — | 68.1 |
| SpreadsheetBench 2 | — | — | — | — | 33.6 |
| HellaSwag | — | — | 85.2 | — | — |
| HLE (with tools) | — | — | — | 60 | 60.9 |
| CharXiv (RQ) | 82.5 | — | — | — | 83.7 |
| WorkIndex (Tencent internal) | — | — | 42.1 | — | — |
| GDP.pdf | 20.8 | — | — | — | 21.6 |
| AIME 2024 | — | 97.1 | — | — | — |
| Belebele (5-language subset) | — | — | 93.8 | — | — |
| KOR-Bench | — | — | 54.7 | — | — |
| CMMLU | — | — | 91.6 | — | — |
| CritPt (no tools) | — | — | — | 18 | 29.1 |
| BrowseComp | — | — | — | 83.4 | 90.8 |
| MMLU-Redux | — | — | 87.6 | — | — |
| NL2Repo-Bench | 61.2 | — | — | 35.5 | 75.3 |
| DRACO | — | — | — | 77.3 | 88.6 |
| SkillsBench | — | — | — | — | 63.7 |
| WebArena-Verified | — | — | — | — | 69.1 |
| SWE-bench Multilingual | — | — | — | 77.3 | 89.5 |
| MMLU | — | 89.6 | 88.7 | — | — |
| Harvey Legal Agent | — | — | — | — | 22.7 |
| PaperBench | — | — | — | — | 86.9 |
| YC-Bench | — | — | — | 500000 | — |
| Agents' Last Exam (CLI) | — | — | — | — | 29.5 |
| Hy-FinAgentBench (Internal) | — | — | — | 78.5 | 82 |
| IFBench | — | 79.2 | — | — | — |
| SWE-bench Verified | — | 76.5 | — | 80.6 | — |
| Humanity's Last Exam | 43.4 | — | — | 42.7 | 54.9 |
| U-MATH | — | — | 69.8 | — | — |
| SpreadSheetBench-v1 | — | — | — | 84.9 | — |
| SUPERChem | — | — | — | 62 | 76.7 |
| VIBE-V2 | — | — | — | 45 | — |
| HorizonMath (pass@4) | — | — | — | 4.42 | 5.3 |
| Terminal-Bench 2.1 (Terminus-2) | — | 80.5 | — | 87.9 | 86.7 |
| DeepSWE | — | — | — | 62.7 | 68.8 |
| BrokenArXiv | — | — | — | 43.1 | 77.7 |
| HMMT Feb 26 | — | 90.5 | — | — | — |
| SocialIQA | — | — | 80.1 | — | — |
| Hy-FinmodelBench v2 (Internal) | — | — | — | 51.3 | 66 |
| MathArena Apex 2025 | — | — | — | 66.3 | 91.4 |
| Terminal Bench 2.1 | — | 80.5 | — | 59.6 | — |
| AIME 2025 | — | 98.3 | — | — | — |
| JobBench | — | — | — | 54.1 | 68 |
| $OneMillion-Bench | — | — | — | 62 | 68.1 |
| OSWorld 2.0 (Binary) | 61.7 | — | — | — | 70.6 |
| C-Eval | — | — | 92 | — | — |
| WideSearch | — | — | — | 81.8 | 84 |
| SimpleVQA | — | — | 48.5 | — | — |
| SkillsBench Avg5 | — | — | — | 65 | 63.7 |
| Vision2Web | — | — | — | — | 80.2 |
| TauBench V3 Banking | — | 31.27 | — | — | 48.7 |
| E-Bench (Internal) | — | — | — | 61.3 | 77.8 |
| ProgramBench | — | — | — | 15.5 | 39.5 |
| MCP-Atlas | — | 71.67 | — | 73.6 | 85.7 |
| SWE-Marathon | — | — | — | 19 | 50 |
| Cybergym | 82.2 | — | — | 83.3 | — |
| Terminal-Bench 4.0 | 38.1 | — | — | — | 51.8 |
| SWE Atlas - RF | — | — | — | 48.6 | 60 |
| DeepSWE 1.1 | 73.7 | 44.3 | — | — | 74 |
| FORTRESS Benign | — | 95.5 | — | — | — |
| Hy-CompanyBench V2 (Internal) | — | — | — | 64.4 | 72.7 |
| C³ | — | — | 77.3 | — | — |
| Harbor-Index | — | — | — | 36.9 | 56.9 |
| SWE Refactor Bench | 23 | — | — | — | 31 |
| SciCode | — | 39.4 | — | — | — |
| AGIEval (English) | — | — | 75.4 | — | — |
| GDPVal-AA v2 | — | — | — | 1580 | 1831 |
| WeaveBench | — | — | — | — | 86.4 |
| SimpleQA | — | — | 37.5 | — | — |
| ZEROBench | — | — | — | — | 52 |
| Covered | 16/125 | 21/125 | 29/125 | 56/125 | 76/125 |
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 | 85.6 | — | 76 of 125 (61.0%) | Vision & Language | Long Context |
| DeepSeek V4 Pro | 57.9 | — | 56 of 125 (45.0%) | General Capabilities | STEM & Reasoning |
| Hy4 flash | 57.7 | — | 29 of 125 (23.0%) | Knowledge | Knowledge |
| MiniMax-M3.1-Flash | 44.3 | — | 16 of 125 (13.0%) | Long Context | STEM & Reasoning |
| Ripple-65B Preview | 40.7 | — | 21 of 125 (17.0%) | Knowledge | General Agents |
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 | DeepSeek V4 Pro | Hy4 flash | MiniMax-M3.1-Flash | Ripple-65B Preview |
|---|---|---|---|---|---|
| General Agents 36 benchmarks | 95.4 | 64.7 | — | 59.5 | 40.9 |
| Agentic Coding 30 benchmarks | 94.9 | 62.2 | — | 52.1 | 54.9 |
| STEM & Reasoning 16 benchmarks | 100 | 60 | — | 46.5 | 49.8 |
| Knowledge 8 benchmarks | — | — | 71.4 | — | 100 |
| Vision & Language 8 benchmarks | 100 | — | — | 96.2 | — |
| Long Context 3 benchmarks | 70 | — | — | 97.8 | — |
| Finance 2 benchmarks | 100 | 66.3 | — | — | — |
| General Capabilities 2 benchmarks | 100 | 76.8 | — | — | — |
| Document Understanding 1 benchmark | 100 | — | — | 93.2 | — |
Partial coverage benchmarks
Scored by at least one but not all selected models (125) — missing scores show as —
| Benchmark | Category | DeepSeek V4 Pro | Claude Opus 5 | MiniMax-M3.1-Flash | Hy4 flash | Ripple-65B Preview |
|---|---|---|---|---|---|---|
| AGIEval (Chinese) 1/5 models | reasoning | — | — | — | 90.6 | — |
| SVG-Bench 1/5 models | coding_agent | 61.7 | — | — | — | — |
| OSWorld-G 1/5 models | general_agent | — | 76.8 | — | — | — |
| GPQA Diamond 3/5 models | stem_reasoning | 92.8 | 93.7 | — | — | 85.2 |
| CountBenchQA 1/5 models | multimodal | — | — | — | 86.6 | — |
| OCW Courses 1/5 models | stem_reasoning | — | — | — | 64 | — |
| ArXivMath 2/5 models | stem_reasoning | 62.1 | 71.5 | — | — | — |
| Chinese-SimpleQA 1/5 models | knowledge | — | — | — | 74.6 | — |
| Hy-LifeSearch (Internal) 2/5 models | general_agent | 46.9 | 56.1 | — | — | — |
| Automation-Bench 3/5 models | general_agent | 30.4 | 48.7 | 62.8 | — | — |
| SWE-bench Pro 3/5 models | coding_agent | 55.4 | 79.2 | — | — | 61.42 |
| BFCL-V4 1/5 models | general_agent | — | 80.6 | — | — | — |
| RealWorldQA 1/5 models | vision_language | — | — | — | 71.1 | — |
| BioMysteryBench 2/5 models | stem_reasoning | 61.6 | 72.1 | — | — | — |
| AA-LCR 1/5 models | long_context | — | — | — | — | 75 |
| AIME 26 1/5 models | stem_reasoning | — | — | — | — | 95 |
| Agents' Last Exam 3/5 models | general_agent | 21.9 | 25.1 | 29.6 | — | — |
| Toolathlon Verified 2/5 models | general_agent | 74.1 | 76.5 | — | — | — |
| Hy-Backend 2.0 (Internal) 2/5 models | coding_agent | 33.8 | 40.3 | — | — | — |
| Hy-SWE Max Verified (Internal) 2/5 models | coding_agent | 65.7 | 70.1 | — | — | — |
| Hy-BrowseComp-Pro2 (Internal) 2/5 models | general_agent | 46.5 | 61.3 | — | — | — |
| ARC-c 1/5 models | reasoning | — | — | — | 95 | — |
| AutomationBench (Public) 1/5 models | general_agent | — | — | — | — | 36.4 |
| GDPVal 2/5 models | general_agent | 70.3 | 80 | — | — | — |
| OfficeQA Pro 2/5 models | general_agent | 65.4 | 66.9 | — | — | — |
| PIQA 1/5 models | reasoning | — | — | — | 84.4 | — |
| StrongREJECT v2 (adaptive) 1/5 models | safety | — | — | — | — | 97.5 |
| SWE Atlas - TW 2/5 models | coding_agent | 45.6 | 68.7 | — | — | — |
| MLE-bench Lite 1/5 models | coding_agent | — | 88 | — | — | — |
| MBPP+ 1/5 models | code | — | — | — | 73.7 | — |
| LOCA-Bench (256k) 2/5 models | long_context | — | 53.7 | 60.3 | — | — |
| WorkBuddy Blind User Evaluation (satisfaction) 1/5 models | general_capabilities | — | — | — | 2.79 | — |
| SWE-bench Pro v2.0 Hard 1/5 models | coding_agent | — | — | — | — | 79.1 |
| SWE Atlas - QnA 2/5 models | coding_agent | 53.4 | 57.5 | — | — | — |
| ApexBench 1/5 models | multimodal_agent | — | 65.8 | — | — | — |
| TextVQA 1/5 models | multimodal | — | — | — | 73.4 | — |
| NL2Repo 2/5 models | coding_agent | 61.5 | 75.3 | — | — | — |
| MLS-Bench 1/5 models | coding_agent | — | 49.8 | — | — | — |
| MMMU-Pro 1/5 models | vision_language | — | — | — | 52 | — |
| Chartography 1/5 models | multimodal_agent | — | 84 | — | — | — |
| PostTrainBench 3/5 models | coding_agent | 24.5 | 35 | 39.3 | — | — |
| INCLUDE 1/5 models | multilingual | — | — | — | 82.8 | — |
| BabyVision 2/5 models | vision_language | — | 94.1 | 91 | — | — |
| Claw-Eval Avg 1/5 models | coding_agent | 58.4 | — | — | — | — |
| OSWorld V2.1 1/5 models | general_agent | — | 74 | — | — | — |
| Apex-Agents 2/5 models | general_agent | 32.4 | 41.8 | — | — | — |
| WorkSpaceBench 2/5 models | general_agent | 65.4 | 75 | — | — | — |
| MMLU-Pro 2/5 models | knowledge | — | — | — | 71.1 | 84.2 |
| Terminal-Bench Science 0.1 1/5 models | coding_agent | — | — | 3.3 | — | — |
| BankerToolBench 3/5 models | general_agent | 73.1 | 81.9 | 76.2 | — | — |
| SuperGPQA 1/5 models | knowledge | — | — | — | 57.8 | — |
| E-Bench-Code (Internal) 2/5 models | coding_agent | 64.3 | 82.7 | — | — | — |
| OSWorld-Verified 1/5 models | general_agent | — | 86.4 | — | — | — |
| CV-Bench 1/5 models | vision_language | — | — | — | 83.4 | — |
| DeepresearchBII 1/5 models | general_agent | — | 54.1 | — | — | — |
| Workspace-Bench-Lite 1/5 models | general_agent | — | 68.1 | — | — | — |
| SpreadsheetBench 2 1/5 models | general_agent | — | 33.6 | — | — | — |
| HellaSwag 1/5 models | reasoning | — | — | — | 85.2 | — |
| HLE (with tools) 2/5 models | stem_reasoning | 60 | 60.9 | — | — | — |
| CharXiv (RQ) 2/5 models | document_understanding | — | 83.7 | 82.5 | — | — |
| WorkIndex (Tencent internal) 1/5 models | general_agent | — | — | — | 42.1 | — |
| GDP.pdf 2/5 models | long_context | — | 21.6 | 20.8 | — | — |
| AIME 2024 1/5 models | stem_reasoning | — | — | — | — | 97.1 |
| Belebele (5-language subset) 1/5 models | multilingual | — | — | — | 93.8 | — |
| KOR-Bench 1/5 models | reasoning | — | — | — | 54.7 | — |
| CMMLU 1/5 models | knowledge | — | — | — | 91.6 | — |
| CritPt (no tools) 2/5 models | stem_reasoning | 18 | 29.1 | — | — | — |
| BrowseComp 2/5 models | general_agent | 83.4 | 90.8 | — | — | — |
| MMLU-Redux 1/5 models | knowledge | — | — | — | 87.6 | — |
| NL2Repo-Bench 3/5 models | coding_agent | 35.5 | 75.3 | 61.2 | — | — |
| DRACO 2/5 models | general_agent | 77.3 | 88.6 | — | — | — |
| SkillsBench 1/5 models | general_agent | — | 63.7 | — | — | — |
| WebArena-Verified 1/5 models | general_agent | — | 69.1 | — | — | — |
| SWE-bench Multilingual 2/5 models | coding_agent | 77.3 | 89.5 | — | — | — |
| MMLU 2/5 models | knowledge | — | — | — | 88.7 | 89.6 |
| Harvey Legal Agent 1/5 models | general_agent | — | 22.7 | — | — | — |
| PaperBench 1/5 models | coding_agent | — | 86.9 | — | — | — |
| YC-Bench 1/5 models | general_agent | 500000 | — | — | — | — |
| Agents' Last Exam (CLI) 1/5 models | general_agent | — | 29.5 | — | — | — |
| Hy-FinAgentBench (Internal) 2/5 models | domain_finance | 78.5 | 82 | — | — | — |
| IFBench 1/5 models | instruction_following | — | — | — | — | 79.2 |
| SWE-bench Verified 2/5 models | coding_agent | 80.6 | — | — | — | 76.5 |
| Humanity's Last Exam 3/5 models | stem_reasoning | 42.7 | 54.9 | 43.4 | — | — |
| U-MATH 1/5 models | stem_reasoning | — | — | — | 69.8 | — |
| SpreadSheetBench-v1 1/5 models | general_agent | 84.9 | — | — | — | — |
| SUPERChem 2/5 models | stem_reasoning | 62 | 76.7 | — | — | — |
| VIBE-V2 1/5 models | coding_agent | 45 | — | — | — | — |
| HorizonMath (pass@4) 2/5 models | stem_reasoning | 4.42 | 5.3 | — | — | — |
| Terminal-Bench 2.1 (Terminus-2) 3/5 models | coding_agent | 87.9 | 86.7 | — | — | 80.5 |
| DeepSWE 2/5 models | coding_agent | 62.7 | 68.8 | — | — | — |
| BrokenArXiv 2/5 models | stem_reasoning | 43.1 | 77.7 | — | — | — |
| HMMT Feb 26 1/5 models | stem_reasoning | — | — | — | — | 90.5 |
| SocialIQA 1/5 models | reasoning | — | — | — | 80.1 | — |
| Hy-FinmodelBench v2 (Internal) 2/5 models | domain_finance | 51.3 | 66 | — | — | — |
| MathArena Apex 2025 2/5 models | stem_reasoning | 66.3 | 91.4 | — | — | — |
| Terminal Bench 2.1 2/5 models | coding_agent | 59.6 | — | — | — | 80.5 |
| AIME 2025 1/5 models | stem_reasoning | — | — | — | — | 98.3 |
| JobBench 2/5 models | general_agent | 54.1 | 68 | — | — | — |
| $OneMillion-Bench 2/5 models | general_capabilities | 62 | 68.1 | — | — | — |
| OSWorld 2.0 (Binary) 2/5 models | general_agent | — | 70.6 | 61.7 | — | — |
| C-Eval 1/5 models | knowledge | — | — | — | 92 | — |
| WideSearch 2/5 models | general_agent | 81.8 | 84 | — | — | — |
| SimpleVQA 1/5 models | vision_language | — | — | — | 48.5 | — |
| SkillsBench Avg5 2/5 models | coding_agent | 65 | 63.7 | — | — | — |
| Vision2Web 1/5 models | vision_language | — | 80.2 | — | — | — |
| TauBench V3 Banking 2/5 models | general_agent | — | 48.7 | — | — | 31.27 |
| E-Bench (Internal) 2/5 models | general_agent | 61.3 | 77.8 | — | — | — |
| ProgramBench 2/5 models | coding_agent | 15.5 | 39.5 | — | — | — |
| MCP-Atlas 3/5 models | general_agent | 73.6 | 85.7 | — | — | 71.67 |
| SWE-Marathon 2/5 models | coding_agent | 19 | 50 | — | — | — |
| Cybergym 2/5 models | general_agent | 83.3 | — | 82.2 | — | — |
| Terminal-Bench 4.0 2/5 models | coding_agent | — | 51.8 | 38.1 | — | — |
| SWE Atlas - RF 2/5 models | coding_agent | 48.6 | 60 | — | — | — |
| DeepSWE 1.1 3/5 models | coding_agent | — | 74 | 73.7 | — | 44.3 |
| FORTRESS Benign 1/5 models | safety | — | — | — | — | 95.5 |
| Hy-CompanyBench V2 (Internal) 2/5 models | general_agent | 64.4 | 72.7 | — | — | — |
| C³ 1/5 models | reasoning | — | — | — | 77.3 | — |
| Harbor-Index 2/5 models | coding_agent | 36.9 | 56.9 | — | — | — |
| SWE Refactor Bench 2/5 models | coding_agent | — | 31 | 23 | — | — |
| SciCode 1/5 models | — | — | — | — | — | 39.4 |
| AGIEval (English) 1/5 models | reasoning | — | — | — | 75.4 | — |
| GDPVal-AA v2 2/5 models | general_agent | 1580 | 1831 | — | — | — |
| WeaveBench 1/5 models | vision_language | — | 86.4 | — | — | — |
| SimpleQA 1/5 models | knowledge | — | — | — | 37.5 | — |
| ZEROBench 1/5 models | vision_language | — | 52 | — | — | — |
Comparison FAQ
Which model scores highest in this comparison?
Claude Opus 5 scores higher overall across category strength profiles (86 vs 58 for DeepSeek V4 Pro); only 0 benchmarks are directly comparable.
Which model is strongest at Agentic Coding?
Claude Opus 5 leads the Agentic Coding category (strength score 94.9/100 across 22 comparable benchmarks).
How reliable is this comparison verdict?
Only 0 of 125 benchmarks are scored by every selected model — the verdict leans on category strength profiles and is indicative.