Kolibri-1 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 90 vs 36 for Kolibri-1, leading 3 of 3 shared benchmarks.
Kimi K3.1 Pro
90
Kolibri-1
35.7
Based on 3 of 92 benchmarks scored by all 2 selected models.
Benchmark Scores Comparison
Kolibri-1
Kimi K3.1 Pro
Kolibri-1
78.1B params
Kimi K3.1 Pro
2.8T params
Coverage matrix
2 models × 92 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 92 benchmarks each model has scores for.
| Benchmark | Kolibri-1 Aleph Alpha | Kimi K3.1 Pro Moonshot AI |
|---|---|---|
| OmniScience Accuracy | 14.8 | — |
| PerceptionBench | — | 68.3 |
| GPQA Diamond | 84.3 | 94.2 |
| IFBench (prompt loose) | 78.1 | — |
| SealQA (12 distractors, <24k) | 61 | — |
| LiveCodeBench v6 | 85.9 | — |
| Design Bench | — | 957 |
| BFCL-V4 | 61.4 | — |
| BioMysteryBench | — | 80.6 |
| AA-LCR | 68.3 | — |
| AIME 26 | 96 | — |
| BenchCAD w/o tools | — | 72.8 |
| BabyVision w/ python | — | 95.4 |
| Agents' Last Exam | — | 43.2 |
| FRAMES (<24k) | 71.2 | — |
| Kimi Hallu Bench w/o tools | — | 5.4 |
| German Public Sector | 75 | — |
| OmniScience Non-Hallucination | 44 | — |
| Aerospace | 58.9 | — |
| Agentic Wiki QA (DE) | 69.4 | — |
| AIME 2025 (DE) | 87.5 | — |
| Kimi Hallu Bench Hard w/ tools | — | 8.3 |
| ZeroBench_main w/ python (pass@5) | — | 63 |
| Overall (EN) | 75.5 | — |
| OfficeQA Pro | — | 67.4 |
| GDP.pdf w/ python | — | 85.7 |
| BenchCAD w/ tools | — | 92.7 |
| ChartoGraphy w/ python | — | 90.2 |
| MuSiQue (EN) | 77.3 | — |
| SealQA (no distractors, >24k) | 100 | — |
| MMMU-Pro | — | 84.1 |
| Program Bench | — | 31.5 |
| PostTrainBench | — | 36.5 |
| OSWorld V2.1 | — | 79.6 |
| Apex-Agents | — | 69.4 |
| HumanEval+ | 92.7 | — |
| MMLU-ProX | 75.5 | — |
| Terminal-Bench Science 0.1 | — | 38.1 |
| ExploitGym (6h) | — | 20.7 |
| Frontier Finance | — | 58.6 |
| xdev-bench-hard | — | 81 |
| HLE (with tools) | — | 70 |
| CharXiv (RQ) | — | 89 |
| Automation Bench-AA | — | 73.3 |
| GDP.pdf | — | 34.6 |
| BrowseComp | 29.4 | 92.1 |
| BFCL v4 (web search) | 62.5 | — |
| RGB Closed-Book | 51 | — |
| Industrial Drive Technology | 60 | — |
| AA-Omniscience Index (public set) | -32.8 | — |
| Harvey Legal Agent | — | 23.7 |
| OfficeQA Pro v2 | — | 64.4 |
| CharXiv (RQ) w/ python | — | 93.4 |
| Agents' Last Exam (CLI) | — | 44.2 |
| Humanity's Last Exam (DE) | 15.9 | — |
| SWE-bench Verified | 66.4 | — |
| Humanity's Last Exam | 21.5 | 51.1 |
| HANDBOOK.md | — | 44.6 |
| BFCL v4 (non-live AST) | 79.1 | — |
| BFCL v4 (live) | 78.9 | — |
| HLE-Diamond (no tools) | — | 39.4 |
| MMMU-Pro w/ python | — | 90 |
| BFCL v4 (memory) | 62.8 | — |
| AIME 2026 (DE) | 90 | — |
| Automotive Supplier | 99 | — |
| GPQA Diamond (DE) | 81.3 | — |
| ExploitBench | — | 74.7 |
| SQuAD (Utility Accuracy) | 83.4 | — |
| Terminal Bench 2.1 | 27.7 | — |
| AIME 2025 | 96.9 | — |
| BFCL v4 (multi-turn) | 47.5 | — |
| RGB Fact-Check (Error Correction) | 34 | — |
| Kimi Code Bench 3.0 Long-horizon | — | 61.2 |
| Kimi Code Bench V2 | — | 77.1 |
| SQuAD (M/A Grounding Score) | 23.4 | — |
| LongBenchPro | 64.5 | — |
| FRAMES (>24k) | 73 | — |
| Semiconductors | 80.4 | — |
| Honeypot | 80.8 | — |
| DECK-Arena | — | 1456 |
| TauBench V3 Banking | 38.1 | — |
| SWE-Marathon | — | 51.7 |
| Terminal-Bench 4.0 | — | 65.7 |
| DeepSWE 1.1 | — | 75.5 |
| RGB Negative (Abstention) | 85.6 | — |
| Business Operation | — | 37.8 |
| MMLU Pro COT | 80 | — |
| Overall (DE) | 70.8 | — |
| AA-Briefcase v1.1 | — | 1678 |
| SealQA (12 distractors, >24k) | 65.1 | — |
| SealQA (no distractors, <24k) | 80.7 | — |
| GDPval-AA v2.1 | — | 1846 |
| Covered | 49/92 | 46/92 |
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 | 90 | 3 of 3 | 46 of 92 (50.0%) | STEM & Reasoning | General Agents |
| Kolibri-1 | 35.7 | 0 of 3 | 49 of 92 (53.0%) | STEM & Reasoning | 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 | Kolibri-1 |
|---|---|---|
| STEM & Reasoning 14 benchmarks | 100 | 60.4 |
| General Agents 12 benchmarks | 100 | 29.6 |
Fully comparable benchmarks
Scored by all 2 selected models (3)
| Benchmark | Category | Kolibri-1 | Kimi K3.1 Pro |
|---|---|---|---|
| GPQA Diamond | stem_reasoning | 84.3 | 94.2 |
| BrowseComp | general_agent | 29.4 | 92.1 |
| Humanity's Last Exam | stem_reasoning | 21.5 | 51.1 |
Partial coverage benchmarks
Scored by at least one but not all selected models (89) — missing scores show as —
| Benchmark | Category | Kolibri-1 | Kimi K3.1 Pro |
|---|---|---|---|
| OmniScience Accuracy 1/2 models | knowledge | 14.8 | — |
| PerceptionBench 1/2 models | vision_language | — | 68.3 |
| IFBench (prompt loose) 1/2 models | instruction_following | 78.1 | — |
| SealQA (12 distractors, <24k) 1/2 models | agentic | 61 | — |
| LiveCodeBench v6 1/2 models | stem_reasoning | 85.9 | — |
| Design Bench 1/2 models | agentic | — | 957 |
| BFCL-V4 1/2 models | general_agent | 61.4 | — |
| BioMysteryBench 1/2 models | stem_reasoning | — | 80.6 |
| AA-LCR 1/2 models | long_context | 68.3 | — |
| AIME 26 1/2 models | stem_reasoning | 96 | — |
| BenchCAD w/o tools 1/2 models | coding_agent | — | 72.8 |
| BabyVision w/ python 1/2 models | vision_language | — | 95.4 |
| Agents' Last Exam 1/2 models | general_agent | — | 43.2 |
| FRAMES (<24k) 1/2 models | agentic | 71.2 | — |
| Kimi Hallu Bench w/o tools 1/2 models | safety | — | 5.4 |
| German Public Sector 1/2 models | agentic | 75 | — |
| OmniScience Non-Hallucination 1/2 models | knowledge | 44 | — |
| Aerospace 1/2 models | agentic | 58.9 | — |
| Agentic Wiki QA (DE) 1/2 models | agentic | 69.4 | — |
| AIME 2025 (DE) 1/2 models | stem_reasoning | 87.5 | — |
| Kimi Hallu Bench Hard w/ tools 1/2 models | safety | — | 8.3 |
| ZeroBench_main w/ python (pass@5) 1/2 models | vision_language | — | 63 |
| Overall (EN) 1/2 models | stem_reasoning | 75.5 | — |
| OfficeQA Pro 1/2 models | general_agent | — | 67.4 |
| 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 |
| MuSiQue (EN) 1/2 models | agentic | 77.3 | — |
| SealQA (no distractors, >24k) 1/2 models | agentic | 100 | — |
| MMMU-Pro 1/2 models | vision_language | — | 84.1 |
| Program Bench 1/2 models | coding_agent | — | 31.5 |
| PostTrainBench 1/2 models | coding_agent | — | 36.5 |
| OSWorld V2.1 1/2 models | general_agent | — | 79.6 |
| Apex-Agents 1/2 models | general_agent | — | 69.4 |
| HumanEval+ 1/2 models | code | 92.7 | — |
| MMLU-ProX 1/2 models | multilingual | 75.5 | — |
| Terminal-Bench Science 0.1 1/2 models | coding_agent | — | 38.1 |
| ExploitGym (6h) 1/2 models | cybersecurity | — | 20.7 |
| Frontier Finance 1/2 models | domain_finance | — | 58.6 |
| xdev-bench-hard 1/2 models | safety | — | 81 |
| HLE (with tools) 1/2 models | stem_reasoning | — | 70 |
| CharXiv (RQ) 1/2 models | document_understanding | — | 89 |
| Automation Bench-AA 1/2 models | agentic | — | 73.3 |
| GDP.pdf 1/2 models | long_context | — | 34.6 |
| BFCL v4 (web search) 1/2 models | agentic | 62.5 | — |
| RGB Closed-Book 1/2 models | grounding_hallucination | 51 | — |
| Industrial Drive Technology 1/2 models | agentic | 60 | — |
| AA-Omniscience Index (public set) 1/2 models | knowledge | -32.8 | — |
| Harvey Legal Agent 1/2 models | general_agent | — | 23.7 |
| 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 |
| Humanity's Last Exam (DE) 1/2 models | stem_reasoning | 15.9 | — |
| SWE-bench Verified 1/2 models | coding_agent | 66.4 | — |
| HANDBOOK.md 1/2 models | general_agent | — | 44.6 |
| BFCL v4 (non-live AST) 1/2 models | agentic | 79.1 | — |
| BFCL v4 (live) 1/2 models | agentic | 78.9 | — |
| HLE-Diamond (no tools) 1/2 models | stem_reasoning | — | 39.4 |
| MMMU-Pro w/ python 1/2 models | vision_language | — | 90 |
| BFCL v4 (memory) 1/2 models | agentic | 62.8 | — |
| AIME 2026 (DE) 1/2 models | stem_reasoning | 90 | — |
| Automotive Supplier 1/2 models | agentic | 99 | — |
| GPQA Diamond (DE) 1/2 models | stem_reasoning | 81.3 | — |
| ExploitBench 1/2 models | cybersecurity | — | 74.7 |
| SQuAD (Utility Accuracy) 1/2 models | grounding_hallucination | 83.4 | — |
| Terminal Bench 2.1 1/2 models | coding_agent | 27.7 | — |
| AIME 2025 1/2 models | stem_reasoning | 96.9 | — |
| BFCL v4 (multi-turn) 1/2 models | agentic | 47.5 | — |
| RGB Fact-Check (Error Correction) 1/2 models | grounding_hallucination | 34 | — |
| Kimi Code Bench 3.0 Long-horizon 1/2 models | coding_agent | — | 61.2 |
| Kimi Code Bench V2 1/2 models | coding_agent | — | 77.1 |
| SQuAD (M/A Grounding Score) 1/2 models | grounding_hallucination | 23.4 | — |
| LongBenchPro 1/2 models | long_context | 64.5 | — |
| FRAMES (>24k) 1/2 models | agentic | 73 | — |
| Semiconductors 1/2 models | agentic | 80.4 | — |
| Honeypot 1/2 models | agentic | 80.8 | — |
| DECK-Arena 1/2 models | agentic | — | 1456 |
| TauBench V3 Banking 1/2 models | general_agent | 38.1 | — |
| SWE-Marathon 1/2 models | coding_agent | — | 51.7 |
| Terminal-Bench 4.0 1/2 models | coding_agent | — | 65.7 |
| DeepSWE 1.1 1/2 models | coding_agent | — | 75.5 |
| RGB Negative (Abstention) 1/2 models | grounding_hallucination | 85.6 | — |
| Business Operation 1/2 models | agentic | — | 37.8 |
| MMLU Pro COT 1/2 models | knowledge | 80 | — |
| Overall (DE) 1/2 models | stem_reasoning | 70.8 | — |
| AA-Briefcase v1.1 1/2 models | general_agent | — | 1678 |
| SealQA (12 distractors, >24k) 1/2 models | agentic | 65.1 | — |
| SealQA (no distractors, <24k) 1/2 models | agentic | 80.7 | — |
| GDPval-AA v2.1 1/2 models | general_agent | — | 1846 |
Comparison FAQ
Which model is better, Kimi K3.1 Pro or Kolibri-1?
Kimi K3.1 Pro wins this comparison with an overall score of 90 vs 36 for Kolibri-1, leading 3 of 3 shared benchmarks.
Which model is better at STEM & Reasoning, Kimi K3.1 Pro or Kolibri-1?
Kimi K3.1 Pro leads the STEM & Reasoning category (strength score 100/100 across 2 comparable benchmarks).
How reliable is this comparison verdict?
Based on 3 of 92 benchmarks scored by all 2 selected models.