APEX TESTING_
Automated benchmark for agentic AI coding models
by HauhauCS
Models Tested
3
Tasks
58
Total Runs
9
Avg Score
79.6
Capital Spent
$0.01
Top Models
View full leaderboard →| # | Model | Provider | ELO |
|---|---|---|---|
| 1 | Claude Opus 4 6 | Anthropic Sub | 2175 |
| 2 | Gpt 5.2 | OpenAI Sub | 2055 |
| 3 | Qwen3 Coder 30B A3B Instruct [F16] | LM Studio | 270 |
Recent Activity
Qwen3 Coder 30B A3B Instruct [F16]→Debug race condition in worker pool
65.08.7s
Qwen3 Coder 30B A3B Instruct [F16]→Build terminal UI dashboard
45.537.7s
Qwen3 Coder 30B A3B Instruct [F16]→Build REST API from scratch
77.07.3s
Gpt 5.2→Debug race condition in worker pool
83.42m 35s
Claude Opus 4 6→Debug race condition in worker pool
93.51m 29s