Fable-5 vs Kimi K3.1 Pro
New ComparisonScores are fetched live from the current database on every visit — nothing is cached.
Winner
Kimi K3.1 Pro
Kimi K3.1 Pro wins this comparison with an overall score of 72 vs 67 for Fable-5, leading 5 of 9 shared benchmarks.
Kimi K3.1 Pro
71.6
Fable-5
67.1
Based on 9 of 76 benchmarks scored by all 2 selected models.
Benchmark Scores Comparison
Fable-5
Kimi K3.1 Pro
Fable-5
Score: 15.0
Kimi K3.1 Pro
2.8T params
Coverage matrix
2 models × 76 benchmarks — hover a benchmark name for its category
Shading = score relative to the best score among the selected models:
Low
Mid
High
Missing (—)
The totals row shows how many of the 76 benchmarks each model has scores for.
| Benchmark | Kimi K3.1 Pro Moonshot AI | Fable-5 Anthropic |
|---|---|---|
| PerceptionBench | 68.3 | — |
| GPQA Diamond | 94.2 | 92.6 |
| Design Bench | 957 | — |
| Automation-Bench | — | 29.1 |
| SWE-bench Pro | — | 80 |
| BioMysteryBench | 80.6 | — |
| PRBench-Finance | — | 55.8 |
| BenchCAD w/o tools | 72.8 | — |
| BabyVision w/ python | 95.4 | — |
| Agents' Last Exam | 43.2 | 23.8 |
| Toolathlon Verified | — | 77.9 |
| PLawBench | — | 70.2 |
| Kimi Hallu Bench w/o tools | 5.4 | — |
| QwenReactBench | — | 1770 |
| Kimi Hallu Bench Hard w/ tools | 8.3 | — |
| ZeroBench_main w/ python (pass@5) | 63 | — |
| OfficeQA Pro | 67.4 | — |
| QwenSVGBench | — | 1690 |
| GDP.pdf w/ python | 85.7 | — |
| BenchCAD w/ tools | 92.7 | — |
| ChartoGraphy w/ python | 90.2 | — |
| PRBench-Legal | — | 57.6 |
| MMMU-Pro | 84.1 | — |
| Program Bench | 31.5 | — |
| PostTrainBench | 36.5 | 41.8 |
| DSBench-FullStack | — | 77.2 |
| Terminal-Bench 3.0 | — | 33.7 |
| OSWorld V2.1 | 79.6 | — |
| Apex-Agents | 69.4 | — |
| WorkSpaceBench | — | 68.7 |
| Terminal-Bench Science 0.1 | 38.1 | — |
| ExploitGym (6h) | 20.7 | 247 |
| Frontier Finance | 58.6 | — |
| xdev-bench-hard | 81 | — |
| HLE (with tools) | 70 | 64.5 |
| CharXiv (RQ) | 89 | — |
| Automation Bench-AA | 73.3 | — |
| FrontierSWE | — | 88.2 |
| GDP.pdf | 34.6 | — |
| BrowseComp | 92.1 | — |
| SkillsBench | — | 70.9 |
| CoWorkBench | — | 75.9 |
| Harvey Legal Agent | 23.7 | — |
| PaperBench | — | 88.8 |
| OfficeQA Pro v2 | 64.4 | — |
| CharXiv (RQ) w/ python | 93.4 | — |
| Agents' Last Exam (CLI) | 44.2 | — |
| IFBench | — | 63.5 |
| Humanity's Last Exam | 51.1 | 53.3 |
| HANDBOOK.md | 44.6 | — |
| HLE-Diamond (no tools) | 39.4 | — |
| MMMU-Pro w/ python | 90 | — |
| DSBench-Hard | — | 68.3 |
| QwenSWEBench | — | 86.3 |
| ExploitGym (2h) | — | 181 |
| DeepSWE | — | 70 |
| ExploitBench | 74.7 | 78 |
| Terminal Bench 2.1 | — | 84.6 |
| Kimi Code Bench 3.0 Long-horizon | 61.2 | — |
| JobBench | — | 57.4 |
| $OneMillion-Bench | — | 55.9 |
| Kimi Code Bench V2 | 77.1 | — |
| QwenQoderBench | — | 63.1 |
| DECK-Arena | 1456 | — |
| WideSearch | — | 81.2 |
| ProgramBench | — | 33 |
| AndroidBench | — | 84.5 |
| SWE-Marathon | 51.7 | 33.1 |
| Cybergym | — | 83.1 |
| Terminal-Bench 4.0 | 65.7 | — |
| DeepSWE 1.1 | 75.5 | 70 |
| Business Operation | 37.8 | — |
| MLS-Bench-Lite | — | 49.9 |
| AA-Briefcase v1.1 | 1678 | — |
| GDPval-AA v2.1 | 1846 | — |
| GDPVal-AA v2 | — | 1743 |
| Covered | 46/76 | 39/76 |
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 |
|---|---|---|---|---|---|
| Kimi K3.1 Pro Winner | 71.6 | 5 of 9 | 46 of 76 (60.0%) | General Agents | Cybersecurity |
| Fable-5 | 67.1 | 4 of 9 | 39 of 76 (51.0%) | Cybersecurity | 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 | Kimi K3.1 Pro | Fable-5 |
|---|---|---|
| Agentic Coding 25 benchmarks | 93.8 | 85.3 |
| General Agents 19 benchmarks | 100 | 40.5 |
| STEM & Reasoning 5 benchmarks | 98.6 | 95.7 |
| Cybersecurity 3 benchmarks | 46.9 | 100 |
Fully comparable benchmarks
Scored by all 2 selected models (9)
| Benchmark | Category | Fable-5 | Kimi K3.1 Pro |
|---|---|---|---|
| GPQA Diamond | stem_reasoning | 92.6 | 94.2 |
| Agents' Last Exam | general_agent | 23.8 | 43.2 |
| PostTrainBench | coding_agent | 41.8 | 36.5 |
| ExploitGym (6h) | cybersecurity | 247 | 20.7 |
| HLE (with tools) | stem_reasoning | 64.5 | 70 |
| Humanity's Last Exam | stem_reasoning | 53.3 | 51.1 |
| ExploitBench | cybersecurity | 78 | 74.7 |
| SWE-Marathon | coding_agent | 33.1 | 51.7 |
| DeepSWE 1.1 | coding_agent | 70 | 75.5 |
Partial coverage benchmarks
Scored by at least one but not all selected models (67) — missing scores show as —
| Benchmark | Category | Fable-5 | Kimi K3.1 Pro |
|---|---|---|---|
| PerceptionBench 1/2 models | vision_language | — | 68.3 |
| Design Bench 1/2 models | agentic | — | 957 |
| Automation-Bench 1/2 models | general_agent | 29.1 | — |
| SWE-bench Pro 1/2 models | coding_agent | 80 | — |
| BioMysteryBench 1/2 models | stem_reasoning | — | 80.6 |
| PRBench-Finance 1/2 models | general_capabilities | 55.8 | — |
| BenchCAD w/o tools 1/2 models | coding_agent | — | 72.8 |
| BabyVision w/ python 1/2 models | vision_language | — | 95.4 |
| Toolathlon Verified 1/2 models | general_agent | 77.9 | — |
| PLawBench 1/2 models | general_capabilities | 70.2 | — |
| Kimi Hallu Bench w/o tools 1/2 models | safety | — | 5.4 |
| QwenReactBench 1/2 models | coding_agent | 1770 | — |
| Kimi Hallu Bench Hard w/ tools 1/2 models | safety | — | 8.3 |
| ZeroBench_main w/ python (pass@5) 1/2 models | vision_language | — | 63 |
| OfficeQA Pro 1/2 models | general_agent | — | 67.4 |
| QwenSVGBench 1/2 models | coding_agent | 1690 | — |
| GDP.pdf w/ python 1/2 models | document_understanding | — | 85.7 |
| BenchCAD w/ tools 1/2 models | coding_agent | — | 92.7 |
| ChartoGraphy w/ python 1/2 models | vision_language | — | 90.2 |
| PRBench-Legal 1/2 models | general_capabilities | 57.6 | — |
| MMMU-Pro 1/2 models | vision_language | — | 84.1 |
| Program Bench 1/2 models | coding_agent | — | 31.5 |
| DSBench-FullStack 1/2 models | coding_agent | 77.2 | — |
| Terminal-Bench 3.0 1/2 models | coding_agent | 33.7 | — |
| OSWorld V2.1 1/2 models | general_agent | — | 79.6 |
| Apex-Agents 1/2 models | general_agent | — | 69.4 |
| WorkSpaceBench 1/2 models | general_agent | 68.7 | — |
| Terminal-Bench Science 0.1 1/2 models | coding_agent | — | 38.1 |
| Frontier Finance 1/2 models | domain_finance | — | 58.6 |
| xdev-bench-hard 1/2 models | safety | — | 81 |
| CharXiv (RQ) 1/2 models | document_understanding | — | 89 |
| Automation Bench-AA 1/2 models | agentic | — | 73.3 |
| FrontierSWE 1/2 models | coding_agent | 88.2 | — |
| GDP.pdf 1/2 models | long_context | — | 34.6 |
| BrowseComp 1/2 models | general_agent | — | 92.1 |
| SkillsBench 1/2 models | general_agent | 70.9 | — |
| CoWorkBench 1/2 models | general_agent | 75.9 | — |
| Harvey Legal Agent 1/2 models | general_agent | — | 23.7 |
| PaperBench 1/2 models | coding_agent | 88.8 | — |
| OfficeQA Pro v2 1/2 models | document_understanding | — | 64.4 |
| CharXiv (RQ) w/ python 1/2 models | vision_language | — | 93.4 |
| Agents' Last Exam (CLI) 1/2 models | general_agent | — | 44.2 |
| IFBench 1/2 models | instruction_following | 63.5 | — |
| HANDBOOK.md 1/2 models | general_agent | — | 44.6 |
| HLE-Diamond (no tools) 1/2 models | stem_reasoning | — | 39.4 |
| MMMU-Pro w/ python 1/2 models | vision_language | — | 90 |
| DSBench-Hard 1/2 models | coding_agent | 68.3 | — |
| QwenSWEBench 1/2 models | coding_agent | 86.3 | — |
| ExploitGym (2h) 1/2 models | cybersecurity | 181 | — |
| DeepSWE 1/2 models | coding_agent | 70 | — |
| Terminal Bench 2.1 1/2 models | coding_agent | 84.6 | — |
| Kimi Code Bench 3.0 Long-horizon 1/2 models | coding_agent | — | 61.2 |
| JobBench 1/2 models | general_agent | 57.4 | — |
| $OneMillion-Bench 1/2 models | general_capabilities | 55.9 | — |
| Kimi Code Bench V2 1/2 models | coding_agent | — | 77.1 |
| QwenQoderBench 1/2 models | coding_agent | 63.1 | — |
| DECK-Arena 1/2 models | agentic | — | 1456 |
| WideSearch 1/2 models | general_agent | 81.2 | — |
| ProgramBench 1/2 models | coding_agent | 33 | — |
| AndroidBench 1/2 models | coding_agent | 84.5 | — |
| Cybergym 1/2 models | general_agent | 83.1 | — |
| Terminal-Bench 4.0 1/2 models | coding_agent | — | 65.7 |
| Business Operation 1/2 models | agentic | — | 37.8 |
| MLS-Bench-Lite 1/2 models | coding_agent | 49.9 | — |
| AA-Briefcase v1.1 1/2 models | general_agent | — | 1678 |
| GDPval-AA v2.1 1/2 models | general_agent | — | 1846 |
| GDPVal-AA v2 1/2 models | general_agent | 1743 | — |
Comparison FAQ
Which model is better, Kimi K3.1 Pro or Fable-5?
Kimi K3.1 Pro wins this comparison with an overall score of 72 vs 67 for Fable-5, leading 5 of 9 shared benchmarks.
Which model is better at Agentic Coding, Kimi K3.1 Pro or Fable-5?
Kimi K3.1 Pro leads the Agentic Coding category (strength score 93.8/100 across 3 comparable benchmarks).
Where does Fable-5 outperform Kimi K3.1 Pro?
Fable-5 leads the Cybersecurity category (strength score 100/100 across 2 comparable benchmarks).
How reliable is this comparison verdict?
Based on 9 of 76 benchmarks scored by all 2 selected models.