Aiera Leaderboard - Grounding: leaderboard
Aiera Leaderboard Grounding axis: whether the claims in a model's financial research answer are tied to explicit sources such as filings, transcripts or research retrieved through Aiera's MCP server, rather than asserted bare. 0-100, higher is better.
Metric: Grounding axis (0-100). Source: huggingface.co. Saturation forecast: Around February 2027. 40 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 97.1 | #20 |
| 2 | Claude Opus 4.7 | 96.2 | #51 |
| 3 | Kimi K3 | 96 | #35 |
| 4 | Kimi K2.7 Code | 95.6 | #95 |
| 5 | Gemini 3.5 Flash | 95.4 | #66 |
| 6 | GPT-5.5 | 95.3 | #33 |
| 7 | Claude Fable 5.1 | 95.3 | #10 |
| 8 | GPT-5.6 Terra | 94.6 | #40 |
| 9 | Inkling | 94.2 | #144 |
| 10 | GPT-5.6 Sol | 93.7 | #22 |
| 11 | Claude Opus 4.8 | 93.7 | #39 |
| 12 | GLM-4.6 | 93.5 | #265 |
| 13 | MiniMax-M3 | 92.9 | #151 |
| 14 | Kimi K2.6 | 92.3 | #113 |
| 15 | DeepSeek V4 Pro | 92.1 | #112 |
No result here: #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=aiera-leaderboard-grounding · How It Works · Data refreshed daily, snapshot 2026-11-05.