Agentic & Tool Use benchmark
PostTrainBench leaderboard
As of October 2026, Claude Fable 5 has the highest published PostTrainBench score on Noometry at 41.8%, out of 11 models with results.
Last verified
About PostTrainBench
A description with primary sources is being prepared for this benchmark.
- Category
- Agentic & Tool Use
- Introduced
- 2026
- Unit
- Percent (random guessing ≈ 0%)
- Official site
- epoch.ai
Top 11 models
- Claude Fable 5 41.8%
- GPT-5.6 Sol 36.2%
- Claude Opus 5 35%
- Claude Opus 4.8 33.8%
- Kimi K3 32%
- GLM-5.2 31.7%
- Claude Opus 4.7 28.6%
- GPT-5.5 27.2%
- Grok 4.5 23.4%
- Gemini 3.1 Pro Preview 22%
- GPT-5.4 19%
Sponsored placements are available on pages like this one. Advertise on Noometry
All results
| # | Model | Provider | Score | Setting | Source | Date |
|---|---|---|---|---|---|---|
| 1 | Claude Fable 5 | Anthropic | 41.8% | max | Epoch AI | |
| 2 | GPT-5.6 Sol | OpenAI | 36.2% | max | Epoch AI | |
| 3 | Claude Opus 5 | Anthropic | 35% | Epoch AI | ||
| 4 | Claude Opus 4.8 | Anthropic | 33.8% | high | Epoch AI | |
| 5 | Kimi K3 | Moonshot AI | 32% | Epoch AI | ||
| 6 | GLM-5.2 | Z.ai (Zhipu) | 31.7% | max | Epoch AI | |
| 7 | Claude Opus 4.7 | Anthropic | 28.6% | xhigh | Epoch AI | |
| 8 | GPT-5.5 | OpenAI | 27.2% | xhigh | Epoch AI | |
| 9 | Grok 4.5 | xAI | 23.4% | high | Epoch AI | |
| 10 | Gemini 3.1 Pro Preview |  Google | 22% | Epoch AI | ||
| 11 | GPT-5.4 | OpenAI | 19% | high | Epoch AI |
Compare the leaders
Frequently asked questions
Which model has the highest PostTrainBench score?
As of October 2026, Claude Fable 5 has the highest published PostTrainBench score on Noometry at 41.8%, out of 11 models with results.
What is the best open-weight model on PostTrainBench?
Kimi K3 has the highest PostTrainBench accuracy among open-weight models at 32%, ranking 5 of 11 overall.