Raon-1 mini vs Qwen3.8-27B
New ComparisonScores are fetched live from the current database on every visit — nothing is cached.
Benchmark Scores Comparison
Raon-1 mini
Qwen3.8-27B
Raon-1 mini
28.8B params
Qwen3.8-27B
27.0B params
Score: 100.0
Coverage matrix
2 models × 50 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 50 benchmarks each model has scores for.
| Benchmark | Qwen3.8-27B Qwen Team (Alibaba Cloud) | Raon-1 mini KRAFTON AI |
|---|---|---|
| GPQA Diamond | 89.2 | 84.3 |
| IFBench (prompt loose) | 79.5 | — |
| LiveCodeBench v6 | 90.3 | 81.6 |
| SWE-bench Pro | 61.7 | — |
| BFCL-V4 | 72.4 | 59.8 |
| RealWorldQA | 85.9 | — |
| Agents' Last Exam (Score) | 42.9 | — |
| AA-LCR | 80.3 | 67.3 |
| MathVision | 90 | — |
| AIME 26 | 98.8 | 93.3 |
| AndroidWorld | 81.9 | — |
| Agents' Last Exam | 20.4 | — |
| Toolathlon Verified | 67.1 | — |
| RecreationBench | 47.1 | — |
| AutomationBench (Public) | 58.7 | 29.7 |
| GDPVal | 63.5 | 66 |
| NL2Repo | 42.3 | — |
| BabyVision | 65.7 | — |
| Claw-Eval Avg | 56.9 | — |
| Multi-Challenge | 63.6 | 59 |
| SWE-MM | 38.6 | — |
| MMLU-Pro | 85.7 | 76.5 |
| OSWorld-Verified | 84.3 | — |
| HLE (with tools) | 30.8 | — |
| LVBench | 72.4 | — |
| CharXiv (RQ) | 83.7 | — |
| GDP.pdf | 11.2 | 5.8 |
| NL2Repo-Bench | 42.3 | — |
| WebArena-Verified | 64.8 | — |
| CoWorkBench | 70.7 | — |
| AA-Omniscience Index (public set) | -0.5 | 3.5 |
| SWE-bench Multilingual | 73.8 | — |
| IFBench | 79.5 | 84.2 |
| SWE-bench Verified | 74.6 | 70.8 |
| Humanity's Last Exam | 30.8 | — |
| ERQA | 65.5 | — |
| QwenSWEBench | 79 | — |
| Terminal-Bench 2.1 (Terminus-2) | 73 | — |
| OmniDocBench 1.5 | 91.1 | — |
| HMMT Feb 26 | 95.8 | 84.1 |
| Terminal-Bench Lite | 67 | 47 |
| JobBench | 33.4 | — |
| OSWorld 2.0 (Partial) | 48 | — |
| OSWorld 2.0 (Binary) | 19.4 | — |
| Vision2Web | 62.9 | — |
| Claw-Eval Pass^3 | 57.4 | — |
| DeepSWE 1.1 | 42.2 | — |
| Chatbot Korean (internal) | 90.1 | 92.9 |
| SciCode | 44.6 | 38.7 |
| BrowseComp Plus | 58 | 55.7 |
| Covered | 50/50 | 18/50 |
Fully comparable benchmarks
Scored by all 2 selected models (18)
| Benchmark | Category | Raon-1 mini | Qwen3.8-27B |
|---|---|---|---|
| GPQA Diamond | stem_reasoning | 84.3 | 89.2 |
| LiveCodeBench v6 | stem_reasoning | 81.6 | 90.3 |
| BFCL-V4 | general_agent | 59.8 | 72.4 |
| AA-LCR | long_context | 67.3 | 80.3 |
| AIME 26 | stem_reasoning | 93.3 | 98.8 |
| AutomationBench (Public) | general_agent | 29.7 | 58.7 |
| GDPVal | general_agent | 66 | 63.5 |
| Multi-Challenge | instruction_following | 59 | 63.6 |
| MMLU-Pro | knowledge | 76.5 | 85.7 |
| GDP.pdf | long_context | 5.8 | 11.2 |
| AA-Omniscience Index (public set) | knowledge | 3.5 | -0.5 |
| IFBench | instruction_following | 84.2 | 79.5 |
| SWE-bench Verified | coding_agent | 70.8 | 74.6 |
| HMMT Feb 26 | stem_reasoning | 84.1 | 95.8 |
| Terminal-Bench Lite | coding_agent | 47 | 67 |
| Chatbot Korean (internal) | multilingual | 92.9 | 90.1 |
| SciCode | — | 38.7 | 44.6 |
| BrowseComp Plus | general_agent | 55.7 | 58 |
Partial coverage benchmarks
Scored by at least one but not all selected models (32) — missing scores show as —
| Benchmark | Category | Raon-1 mini | Qwen3.8-27B |
|---|---|---|---|
| IFBench (prompt loose) 1/2 models | instruction_following | — | 79.5 |
| SWE-bench Pro 1/2 models | coding_agent | — | 61.7 |
| RealWorldQA 1/2 models | vision_language | — | 85.9 |
| Agents' Last Exam (Score) 1/2 models | general_agent | — | 42.9 |
| MathVision 1/2 models | vision_language | — | 90 |
| AndroidWorld 1/2 models | general_agent | — | 81.9 |
| Agents' Last Exam 1/2 models | general_agent | — | 20.4 |
| Toolathlon Verified 1/2 models | general_agent | — | 67.1 |
| RecreationBench 1/2 models | general_agent | — | 47.1 |
| NL2Repo 1/2 models | coding_agent | — | 42.3 |
| BabyVision 1/2 models | vision_language | — | 65.7 |
| Claw-Eval Avg 1/2 models | coding_agent | — | 56.9 |
| SWE-MM 1/2 models | coding_agent | — | 38.6 |
| OSWorld-Verified 1/2 models | general_agent | — | 84.3 |
| HLE (with tools) 1/2 models | stem_reasoning | — | 30.8 |
| LVBench 1/2 models | video_understanding | — | 72.4 |
| CharXiv (RQ) 1/2 models | document_understanding | — | 83.7 |
| NL2Repo-Bench 1/2 models | coding_agent | — | 42.3 |
| WebArena-Verified 1/2 models | general_agent | — | 64.8 |
| CoWorkBench 1/2 models | general_agent | — | 70.7 |
| SWE-bench Multilingual 1/2 models | coding_agent | — | 73.8 |
| Humanity's Last Exam 1/2 models | stem_reasoning | — | 30.8 |
| ERQA 1/2 models | spatial_intelligence | — | 65.5 |
| QwenSWEBench 1/2 models | coding_agent | — | 79 |
| Terminal-Bench 2.1 (Terminus-2) 1/2 models | coding_agent | — | 73 |
| OmniDocBench 1.5 1/2 models | document_understanding | — | 91.1 |
| JobBench 1/2 models | general_agent | — | 33.4 |
| OSWorld 2.0 (Partial) 1/2 models | general_agent | — | 48 |
| OSWorld 2.0 (Binary) 1/2 models | general_agent | — | 19.4 |
| Vision2Web 1/2 models | vision_language | — | 62.9 |
| Claw-Eval Pass^3 1/2 models | coding_agent | — | 57.4 |
| DeepSWE 1.1 1/2 models | coding_agent | — | 42.2 |