DTeval

A Collective Intelligence Project

Loading blueprint versions...

Please wait while we gather all the unique runs for this blueprint.

Analysis: Gd7 Experiment Dtef Global Dialogues Gd7 Religion%3Aislam B2 - Run 65146ab...

Global Dialogues GD7 - Islam (batch 2)

DTEF: Batched predictions (2 Qs/prompt) for Islam. Source: Global Dialogues GD7 (https://github.com/collect-intel/global-dialogues)

TAGS:

Global Dialogues Gd7

Instruction Following & Prompt Adherence

Factual Accuracy & Hallucination

AI Safety & Robustness

Best Models (Coverage)

1.Gemini 2.5 Flash
81.3%
2.Claude Haiku 4.5
80.4%
3.GPT 4o Mini
80.3%
4.GPT 4.1 Mini
78.5%
5.GPT 4.1 Nano
78.3%

Select Prompt:

Macro Coverage Overview

Average key point coverage extent for each model across all prompts.

Pro Tip

Click on any result cell to open a detailed view.

Advanced view

Highlight best performers

Sort prompts by

Sort models by

Color Scale - Simplified View (Avg. Coverage)

Perfect

Excellent

Good

Fair

Poor

Bad

Not Met

	Prompts vs. Models	Claude Haiku 4.5	Gemini 2.5 Flash	Gemma 3 12b It	GPT 4.1 Mini	GPT 4.1 Nano	GPT 4o Mini
Score		2nd 80.4%	1st 81.3%	6th 72.5%	4th 78.5%	5th 78.3%	3rd 80.3%
80.0%		84%	86%	69%	83%	79%	79%
83.0%		85%	86%	82%	83%	74%	88%
74.2%		75%	73%	58%	77%	85%	77%
73.8%		77%	81%	62%	73%	74%	76%
83.5%		78%	86%	86%	83%	86%	82%
79.8%		85%	80%	75%	82%	77%	80%
84.5%		83%	84%	83%	83%	84%	90%
78.0%		80%	83%	80%	72%	72%	81%
69.0%		74%	74%	63%	65%	71%	67%
79.7%		83%	80%	67%	84%	81%	83%