Coding and agents
Task-specific coding and agentic benchmark leaders.
Snapshot: August 4, 2026
Agentic
Artificial Analysis Agentic Index · Higher is better
- Eligible
- 176
- Candidates
- 591
- Coverage
- 29.8%
| Rank | Model | Creator | Artificial Analysis Agentic Index | Since prior snapshot |
|---|---|---|---|---|
| 1 | Claude Opus 5 (Adaptive Reasoning, Max Effort) | Anthropic | 55.3 | No change |
| 2 | Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) | Anthropic | 54.5 | No change |
| 3 | GPT-5.6 Sol (max) | OpenAI | 54 | No change |
Coding
Artificial Analysis Coding Index · Higher is better
- Eligible
- 203
- Candidates
- 591
- Coverage
- 34.3%
| Rank | Model | Creator | Artificial Analysis Coding Index | Since prior snapshot |
|---|---|---|---|---|
| 1 | GPT-5.6 Sol (xhigh) | OpenAI | 78.3 | No change |
| 2 | Claude Opus 5 (Adaptive Reasoning, Max Effort) | Anthropic | 78 | No change |
| 3 | GPT-5.6 Sol (max) | OpenAI | 77.4 | No change |
Data source: Artificial Analysis .