Best LLM for structured outputs (JSON)
- models ranked
- 183
- lead over #2
- +1.6
Overall index · read October 21
# Overall index, read October 21
leaders = {
"Claude Opus 5.5": 57.6, "Claude Sonnet 5.5": 56.0, "Claude Fable 5.1": 53.4,
}
As of October 21, 2026, Anthropic Claude Opus 5.5 leads 183 models that qualify, with 57.6 on the overall index, ahead of Anthropic Claude Sonnet 5.5 (56.0) and Anthropic Claude Fable 5.1 (53.4).
| Rank | Model | Overall index | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 Anthropic | 57.6 | 92.2 | $20.00 | |
| 2 | Claude Sonnet 5.5 Anthropic | 56.0 | 88.9 | $10.00 | |
| 3 | Claude Fable 5.1 Anthropic | 53.4 | 89.1 | $50.00 | |
| 4 | GPT-6 Astra OpenAI | 52.7 | 88.7 | $50.00 | |
| 5 | Gemini 4 Argon Google | 52.6 | 80.5 | $10.00 | |
| 6 | GPT-6.1 Sol OpenAI | 51.8 | 88.7 | $10.00 | |
| 7 | Claude Opus 5 Anthropic | 50.8 | 81.8 | $25.00 | |
| 8 | Claude Fable 5 Anthropic | 49.6 | 82.1 | $50.00 | |
| 9 | Muse Spark 1.3 Meta | 48.1 | 79.7 | $4.25 | |
| 10 | GPT-6 Sol OpenAI | 47.6 | 82.8 | $10.00 | |
| 11 | GPT-5.6 Sol OpenAI | 47.0 | 76.7 | $20.00 | |
| 12 | Grok 4.7 xAI | 46.4 | 75.0 | $6.00 | |
| 13 | MiMo-V2.6-Pro Xiaomi | 46.3 | 72.6 | $0.87 | |
| 14 | Qwen3.8 Max (0902) Qwen | 45.4 | 63.5 | $6.00 | |
| 15 | GLM 5.3 Z.ai | 44.8 | 73.1 | $1.40 | |
| 16 | Grok 4.6 xAI | 44.3 | 76.4 | $6.00 | |
| 17 | Step 5 Preview StepFun | 43.7 | 71.4 | $2.70 | |
| 18 | Kimi K3 Moonshot AI | 43.6 | 61.8 | $11.936 | |
| 19 | Claude Haiku 5.5 Anthropic | 43.4 | 75.9 | $0.50 | |
| 20 | GPT-5.6 Terra OpenAI | 42.1 | 69.0 | $12.00 |
How this list is built
Generated from the dataset on every refresh. Nobody picks the order by hand.
Models whose OpenRouter listing supports structured outputs, ordered by the Artificial Analysis Intelligence Index, highest first. The index is a composite of evaluations Artificial Analysis runs itself, published on a 0 to 100 scale.
- OpenRouter, fetched
- Artificial Analysis (via OpenRouter), fetched