TelcoAgent-Bench - Resolution Accuracy (Arabic): leaderboard

TelcoAgent-Bench (2026) tests small language models as telecom troubleshooting agents: from the first message of an engineer the agent recognizes the intent, calls network tools along a reference path among distractor tools and writes a resolution summary. Resolution accuracy as cosine similarity times 100 between the final summary and the gold summary in Arabic dialogues; higher is better.

Metric: Resolution accuracy: mean cosine similarity x100 (-100 to 100, sentence embeddings) between the agent's final resolution summary and the gold summary on TelcoAgent-Bench (15 telecom troubleshooting intents, 49 blueprints with 30 sampled dialogues each; the agent starts from the engineer's first message and calls core and distractor network tools), Arabic dialogues; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 8 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemma 3 4B (IT)84.7#4139
2Qwen 3 8B82.8#1873
3Granite 3.3 8B Instruct82.6#3866
4Llama 3 8B Instruct82.4#4852
5Mistral 7B Instruct81.7#7568
6Qwen 2.5 7B Instruct80.6#2723
7granite-3.1-3B-a800m-instruct78.9#8600
8Qwen-7B Chat73.3#7138

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=telcoagent-bench-resolution-accuracy-arabic · How It Works · Data refreshed daily, snapshot 2026-11-21.