Telco-GAIA: leaderboard
Telecommunications questions solved by agents using tools and external information. Reports the publisher’s answer accuracy; cost and elapsed time are separate operational measurements. This board reports Telco-GAIA using Accuracy (%).
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around February 2027. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.8 | 71 | #39 |
| 2 | GPT-5.5 | 68 | #33 |
| 3 | Kimi K2.6 | 68 | #113 |
| 4 | Claude Sonnet 4.6 | 67 | #94 |
| 5 | Gemini 3.5 Flash | 67 | #66 |
| 6 | DeepSeek V4 Pro | 48 | #112 |
| 7 | GPT-5.4 | 42 | #83 |
| 8 | DeepSeek V4 Flash | 40 | #155 |
| 9 | GPT-5.2 | 38 | #115 |
| 10 | Gemini 3.1 Flash Lite | 37 | #200 |
| 11 | GPT-5.4 Mini | 15 | #228 |
| 12 | GPT-5.4 Nano | 13 | #359 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=telco-gaia · How It Works · Data refreshed daily, snapshot 2026-11-05.