Taplio - Build a growth routine on LinkedIn
Mini toolsAider Polyglot
Aider Polyglot Coding Benchmark
225 hard Exercism programming exercises across 6 languages (C++, Go, Java, JavaScript, Python, Rust). Measures whole-file edit accuracy under realistic agentic-coding harness.
Coding
Text
accuracy
Max 100.0%
Released Dec 2024
12
Results
12
Models scored
89.4%
Top: Claude Opus 4.5
71.5%
Median
Best results
Top primary scores; one row per model.
Frontier over time
Each dot is one model result; the line traces the running best score.
All results
Showing one canonical row per model. Show all configurations
| # | Model | Score | Conditions | Eval date | Source | Flags |
|---|---|---|---|---|---|---|
| 1 | Claude Opus 4.5 | 89.4% | — | Nov 24, 2025 | self reported | primary |
| 2 | GPT 5.1 | 88.0% | 0-shot · CoT | Nov 13, 2025 | self reported | primary |
| 3 | GPT 5 (Thinking) | 88.0% | — | Aug 7, 2025 | self reported | primary |
| 4 | o3 (High) | 81.3% | — | Apr 16, 2024 | self reported | primary |
| 5 | Claude Sonnet 4.5 | 78.8% | — | Nov 24, 2025 | self reported | primary |
| 6 | Gemini 2.5 Pro | 74.0% | — | Jun 17, 2025 | third party | primary verified |
| 7 | o4 mini (high) | 68.9% | — | Apr 16, 2025 | self reported | primary |
| 8 | o1 (High) | 64.4% | — | Apr 16, 2025 | self reported | primary |
| 9 | Qwen3 235B A22B | 61.8% | Pass@2 | Apr 28, 2025 | self reported | primary |
| 10 | GPT 4.1 | 52.0% | — | Apr 14, 2025 | self reported | primary |
| 11 | Gemini 2.5 Flash-Lite | 26.7% | — | Sep 26, 2025 | self reported | primary |
| 12 | GPT 5 | 26.7% | — | Aug 7, 2025 | self reported | primary |
