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 Environment%3Arural B2 - Run ac7306a...

Global Dialogues GD7 - Rural (batch 2)

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

TAGS:

Global Dialogues Gd7

Instruction Following & Prompt Adherence

Mathematics & Statistics

General Knowledge

Social interaction

Best Models (Coverage)

1.Gemini 2.5 Flash
80.4%
2.Claude Haiku 4.5
79.3%
3.GPT 4o Mini
78.8%
4.GPT 4.1 Mini
78.2%
5.GPT 4.1 Nano
75.0%

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 79.3%	1st 80.4%	6th 71.1%	4th 78.2%	5th 75.0%	3rd 78.8%
75.3%		75%	85%	72%	75%	70%	75%
83.8%		82%	88%	83%	87%	80%	83%
77.2%		84%	88%	59%	80%	74%	78%
78.8%		82%	85%	73%	73%	77%	83%
83.3%		86%	85%	87%	84%	73%	85%
71.2%		76%	69%	63%	75%	72%	72%
81.3%		78%	78%	81%	81%	83%	87%
78.0%		78%	83%	77%	79%	68%	83%
64.2%		70%	64%	54%	68%	68%	61%
78.2%		82%	79%	62%	80%	85%	81%