DTeval

A Collective Intelligence Project

Loading blueprint versions...

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

Analysis: Gd4 Experiment Dtef Global Dialogues Gd4 Country%3Asouth Africa B2 - Run da15116...

Global Dialogues GD4 - South Africa (batch 2)

DTEF: Batched predictions (2 Qs/prompt) for South Africa. Source: Global Dialogues GD4 (https://github.com/collect-intel/global-dialogues)

TAGS:

Global Dialogues Gd4

Instruction Following & Prompt Adherence

Factual Accuracy & Hallucination

General Knowledge

Economics & Finance

Political Science

Sociology & Anthropology

AI Safety & Robustness

Best Models (Coverage)

1.Claude Haiku 4.5
59.7%
2.Gemini 2.5 Flash
57.2%
3.GPT 4o Mini
55.0%
4.GPT 4.1 Nano
54.8%
5.GPT 4.1 Mini
54.3%

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		1st 59.7%	2nd 57.2%	6th 45.3%	5th 54.3%	4th 54.8%	3rd 55.0%
66.3%		79%	66%	54%	71%	58%	70%
36.7%		41%	44%	21%	35%	49%	30%
54.3%		59%	62%	43%	52%	58%	52%
72.5%		74%	77%	65%	71%	72%	76%
50.3%		57%	49%	44%	49%	50%	53%
57.0%		61%	56%	47%	60%	47%	71%
33.7%		40%	35%	29%	29%	30%	39%
75.0%		79%	79%	74%	76%	71%	71%
46.2%		51%	54%	36%	35%	56%	45%
51.8%		56%	50%	40%	65%	57%	43%