PostTrainBench
Category
Agentic Coding
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for PostTrainBench
| Model | Score | Date | Verified |
|---|---|---|---|
| GPT 5.5 |
89.42%
|
11.06.2026 | Unverified |
| MiniMax-M2.7 |
13.10
|
11.06.2026 | Unverified |
| Claude Opus 4.7 |
100.00%
|
11.06.2026 | Unverified |
| MiniMax-M3 |
81.91%
|
11.06.2026 | Unverified |
| Gemini 3.1 Pro |
7.17%
|
11.06.2026 | Unverified |
| Kimi K3 |
80.20%
|
18.08.2026 | Verified |
| GLM-5.2 |
72.35%
|
17.06.2026 | Verified |
| GLM-5.1 |
23.89%
|
17.06.2026 | Unverified |
| GPT-5.5 |
52.22%
|
17.06.2026 | Unverified |
| Claude Opus 4.8 |
82.25%
|
17.06.2026 | Unverified |
| GLM-5.3 |
91.13%
|
28.08.2026 | Verified |
| Opus-4.8 |
67.58%
|
28.08.2026 | Verified |
| Fable-5 |
97.95%
|
28.08.2026 | Verified |
| GPT-5.6 Sol |
78.84%
|
28.08.2026 | Verified |
| Hy3 |
4.78%
|
28.08.2026 | Verified |
| Hy4 preview |
76.79%
|
28.08.2026 | Verified |
| DeepSeek V4 Pro |
38.91%
|
28.08.2026 | Verified |
| Claude Opus 5 |
74.74%
|
28.08.2026 | Verified |