DTeval

Loading analysis results...

Please wait while we prepare the detailed comparison.

Loading blueprint versions...

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

Analysis: Gd3 C10 Dtef Global Dialogues Gd3 Country%3Abangladesh C10 - Run df38f77...

Global Dialogues GD3 - Bangladesh (10 context Qs)

DTEF: Predict response distributions for Bangladesh. Source: Global Dialogues GD3 (https://github.com/collect-intel/global-dialogues)

TAGS:

Instruction Following & Prompt Adherence

Factual Accuracy & Hallucination

Reasoning

AI Safety & Robustness

Best Models (Coverage)

1.Claude Sonnet 4
74.7%
2.Claude 3.7 Sonnet
74.2%
3.Claude Sonnet 4.5
74.1%
4.GPT 4o
73.9%
5.GPT 5.1
73.7%

🔀 Least Similar Models

Claude Sonnet 4.5vsGPT OSS 20b

66.4% similarity

👯 Most Similar Models

GPT 4.1vsGPT 4o

91.5% similarity

View Blueprint

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

	Claude 3.7 Sonnet	Claude Haiku 4.5	Claude Sonnet 4	Claude Sonnet 4.5	Gemini 2.5 Flash	Gemma 3 12b It	Llama 4 Maverick	Mistral Large 2411	Mistral Medium 3	GPT 4.1	GPT 4.1 Mini	GPT 4.1 Nano	GPT 4o	GPT 4o Mini	GPT 5	GPT 5.1	GPT OSS 120b	GPT OSS 20b	Qwen3 30b A3B Instruct 2507	Qwen3 32b	Grok 4
Score	2nd 74.2%	7th 73.3%	1st 74.7%	3rd 74.1%	19th 65.3%	20th 64.9%	11th 71.5%	10th 72.0%	8th 73.0%	9th 72.3%	14th 70.2%	13th 70.2%	4th 73.9%	17th 66.5%	6th 73.3%	5th 73.7%	12th 71.2%	15th 66.8%	16th 66.8%	18th 66.3%	21st 62.3%
77.8%	77%	85%	71%	75%	67%	69%	77%	78%	84%	78%	81%	79%	84%	81%	90%	88%	82%	61%	82%	82%	63%
78.3%	72%	92%	87%	73%	38%	78%	72%	87%	83%	73%	79%	88%	83%	86%	86%	81%	83%	59%	79%	86%	80%
77.8%	86%	72%	86%	81%	65%	80%	80%	81%	65%	86%	86%	86%	86%	86%	65%	80%	77%	75%	86%	65%	60%
63.6%	66%	62%	62%	67%	73%	61%	62%	53%	58%	68%	68%	73%	68%	58%	63%	65%	65%	62%	52%	73%	57%
78.1%	67%	92%	71%	82%	78%	66%	67%	71%	83%	100%	82%	69%	87%	89%		75%	78%	87%	67%	86%	66%
78.0%	71%	78%	71%	80%	71%	77%	77%	86%	76%	77%	77%	73%	77%	77%	77%	77%	91%	77%	73%	90%	86%
76.3%	79%	76%	79%	79%	82%	77%	72%	75%	74%	80%	76%	74%	76%	77%	87%	76%	77%	79%	72%	80%	56%
81.1%	92%	92%	83%	76%	60%	52%	79%	92%	100%	72%	80%	86%	82%	68%	92%	89%	93%	94%	77%	77%	67%
84.0%	79%	90%	93%	83%	93%	67%	89%	94%	83%	83%	93%	83%	89%	76%	76%	87%	87%	92%	89%	86%	52%
77.5%	92%	77%	92%	83%	73%	73%	74%	67%	73%	81%	73%	73%	73%	73%	85%	76%	75%	76%	73%	77%	88%
81.1%	86%	86%	87%	86%	43%	83%	81%	88%	88%	83%	88%	88%	80%	80%	82%	87%	75%	76%	75%	85%	76%
61.5%	74%	57%	63%	67%	69%	63%	63%	62%	67%	66%	55%	65%	45%	47%	61%	62%	56%	59%	68%	60%	62%
68.0%	61%	84%	70%	70%	80%	80%	73%	64%	76%	67%	80%	60%	81%	79%	67%	64%	48%	48%	79%	48%	48%
82.5%	92%	92%	92%	83%	81%	76%	87%	90%	87%	83%	92%	80%	92%	80%	78%	91%	92%	80%	92%	46%	46%
73.0%	83%	62%	80%	86%	82%	58%	79%	83%	79%	74%	55%	90%	75%	54%	80%	68%	75%	54%	48%	89%	78%
58.8%	60%	65%	70%	60%	52%	45%	63%	67%	67%	63%	46%	57%	70%	53%	64%	56%	60%	69%	67%	49%	32%
54.3%	73%	64%	73%	64%	80%	40%	60%	38%	58%	52%	56%	46%	38%	42%	61%	64%	52%	34%	54%	46%	46%
72.0%	86%	77%	87%	86%	72%	44%	70%	76%	76%	71%	60%	70%	73%	61%	76%	80%	73%	62%	67%	70%	74%
51.7%	44%	42%	52%	40%	49%	54%	57%	65%	58%	62%	42%	28%	65%	53%	65%	71%	64%	32%	42%	45%	55%
73.6%	80%	72%	72%	77%	80%	77%	72%	74%	71%	72%	74%	74%	71%	74%	70%	74%	74%	72%	71%	73%	71%
76.8%	82%	88%	88%	69%	59%	64%	88%	70%	90%	80%	90%	90%	88%	67%	79%	82%	74%	72%	63%	76%	53%
70.3%	64%	69%	67%	67%	47%	62%	77%	77%	82%	75%	66%	71%	88%	66%	71%	82%	75%	78%	62%	73%	58%
69.4%	67%	67%	68%	65%	61%	88%	82%	82%	67%	67%	67%	63%	79%	65%	70%	67%	82%	69%	68%	53%	61%
65.6%	60%	76%	64%	60%	57%	85%	66%	64%	74%	66%	83%	63%	74%	78%	59%	73%	61%	48%	71%	48%	48%
62.7%	64%	64%	61%	61%	55%	64%	65%	65%	62%	62%	62%	61%	67%	61%	67%	62%	66%	63%	65%	61%	59%
55.8%	49%	57%	48%	47%	46%	58%	51%	58%	68%	48%	61%	77%	59%	69%	42%	58%	65%	66%	62%	57%	25%
63.4%	71%	69%	76%	71%	69%	42%	62%	71%	71%	69%	49%	56%	71%	54%	71%	66%	68%	69%	51%	67%	38%
55.7%	74%	53%	52%	74%	90%	34%	53%	63%	63%	59%	43%	49%	58%	33%	67%	56%	54%	52%	32%	46%	64%
67.5%	80%	74%	76%	80%	82%	55%	67%	56%	74%	59%	61%	73%	56%	53%	74%	72%	56%	71%	59%	59%	80%
76.0%	80%	68%	82%	100%	72%	69%	65%	82%	73%	82%	80%	82%	80%	63%	83%	82%	72%	51%	89%	84%	56%
73.9%	93%	75%	86%	86%	64%	59%	82%	87%	76%	75%	61%	68%	87%	60%	85%	87%	79%	55%	53%	71%	62%
73.6%	67%	82%	67%	73%	89%	73%	73%	73%	89%	67%	67%	67%	67%	67%	79%	73%	76%	79%	67%	84%	67%
57.1%	63%	50%	63%	58%	44%	58%	58%	58%	44%	63%	63%	63%	63%	63%	53%	58%	55%	53%	63%	53%	53%
76.5%	82%	71%	82%	82%	65%	74%	77%	74%	83%	79%	74%	74%	74%	74%	82%	80%	77%	84%	65%	74%	79%
63.9%	68%	66%	76%	65%	26%	69%	55%	55%	66%	68%	62%	55%	76%	66%	62%	73%	63%	70%	70%	59%	71%
75.0%	76%	79%	77%	89%	51%	72%	79%	79%	79%	93%	76%	69%	79%	49%	86%	80%	78%	69%	59%	73%	82%
71.2%	78%	78%	76%	84%	79%	56%	78%	79%	71%	68%	57%	66%	78%	57%	76%	71%	71%	64%	68%	59%	81%
80.6%	73%	87%	86%	89%	65%	72%	84%	84%	78%	78%	93%	86%	84%	88%	86%	89%	75%	92%	86%	73%	44%
59.8%	68%	60%	72%	66%	61%	54%	68%	68%	55%	68%	62%	59%	61%	48%	71%	66%	64%	36%	41%	45%	62%
93.2%	100%	85%	100%	94%	78%	94%	94%	94%	78%	100%	100%	100%	100%	100%	89%	94%	91%	89%	100%	100%	78%
64.1%	77%	74%	68%	68%	82%	61%	74%	64%	71%	68%	58%	49%	72%	52%	79%	65%	62%	66%	43%	19%	74%
55.9%	65%	60%	54%	58%	49%	42%	60%	55%	52%	60%	58%	62%	54%	58%	58%	55%	60%	58%	54%	48%	53%
66.7%	68%	82%	80%	82%	31%	67%	63%	45%	66%	66%	82%	74%	68%	73%	66%	66%	59%	72%	67%	56%	67%

Model Similarity Dendrogram

Hierarchical clustering of models based on response similarity. Models grouped closer are more similar.