edge cases
Feb 21, 2026EDGE-006Calculate and explain any issues with: 1. 0.1 + 0.2 = ? 2. 2^53 + 1 in JavaScript 3. 1/3 represented as a finite decimal 4. sqrt(-1) in Python without importing cmath 5. 10^309 in most programming languages 6. What's the result of: (-1) % 10 in Python vs JavaScript? Don't just give answers - explain WHY these are problematic.
Winner
Claude Sonnet 4.5
Anthropic
9.83
WINNER SCORE
matrix avg: 9.35
10×10 Judgment Matrix · 100 judgments
OPEN DATA
| Judge ↓ / Respondent → | Claude Opus 4.5 | Gemini 3 | Claude Sonnet 4.5 | GPT-5.2-Codex | GPT-OSS-120B | Gemini 3 | DeepSeek V3.2 | MiMo-V2-Flash | Grok 4.1 Fast | Grok 3 (Direct) |
|---|---|---|---|---|---|---|---|---|---|---|
| Claude Opus 4.5 | — | 7.5 | 9.8 | 9.3 | 8.8 | 9.8 | 9.4 | 9.6 | 10.0 | 9.8 |
| Gemini 3 | 10.0 | — | 10.0 | 0.0 | 0.0 | 0.0 | 9.8 | 0.0 | 0.0 | 0.0 |
| Claude Sonnet 4.5 | 10.0 | 8.3 | — | 9.8 | 9.8 | 9.8 | 9.0 | 10.0 | 9.8 | 9.8 |
| GPT-5.2-Codex | 9.0 | 5.0 | 9.8 | — | 7.5 | 8.8 | 9.4 | 8.0 | 9.2 | 8.6 |
| GPT-OSS-120B | 0.0 | 0.0 | 9.4 | 9.3 | — | 9.2 | 0.0 | 8.8 | 0.0 | 0.0 |
| Gemini 3 | 10.0 | 8.1 | 10.0 | 9.8 | 9.6 | — | 9.8 | 9.6 | 10.0 | 9.8 |
| DeepSeek V3.2 | 10.0 | 8.8 | 10.0 | 9.1 | 9.2 | 9.8 | — | 9.8 | 9.8 | 9.8 |
| MiMo-V2-Flash | 9.8 | 8.6 | 9.8 | 9.8 | 9.0 | 8.6 | 9.8 | — | 10.0 | 9.0 |
| Grok 4.1 Fast | 10.0 | 7.5 | 10.0 | 10.0 | 9.2 | 10.0 | 9.8 | 10.0 | — | 10.0 |
| Grok 3 (Direct) | 9.7 | 7.7 | 9.6 | 9.4 | 8.8 | 9.4 | 8.8 | 9.7 | 9.7 | — |