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 C5 Dtef Global Dialogues Gd3 Country%3Aspain C5 - Run 057954f...

Global Dialogues GD3 - Spain (5 context Qs)

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

TAGS:

Instruction Following & Prompt Adherence

Reasoning

Efficiency & Succinctness

Factual Accuracy & Hallucination

Best Models (Coverage)

1.Claude Sonnet 4
72.6%
2.Llama 4 Maverick
71.5%
3.GPT 5.1
71.3%
4.Gemini 2.5 Flash
71.2%
5.GPT 4.1
71.0%

🔀 Least Similar Models

Claude Sonnet 4.5vsClaude Sonnet 4

53.1% similarity

👯 Most Similar Models

GPT 4.1 MinivsGPT 4o

95.9% 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	7th 70.3%	16th 66.1%	1st 72.6%	6th 70.3%	4th 71.2%	20th 60.8%	2nd 71.5%	11th 68.7%	13th 67.6%	5th 71.0%	10th 68.8%	21st 58.3%	17th 65.8%	19th 62.5%	8th 69.3%	3rd 71.3%	12th 68.1%	18th 64.1%	14th 66.8%	15th 66.5%	9th 69.2%
73.6%	73%		85%	85%	65%					73%					70%	67%	70%		88%	67%	67%
79.0%	78%									87%						80%					71%
74.8%	79%		71%	83%	75%					75%					71%	73%	75%		67%	83%	71%
68.5%	78%	60%	68%	60%	60%	48%	68%	70%	67%	67%	73%	64%	78%	78%	75%	67%	73%	73%	60%	76%	75%
84.8%	82%		76%	76%	91%					91%					84%	86%	89%		76%	91%	91%
69.5%	69%		60%	60%	65%					75%					65%	65%	76%		75%	80%	75%
60.5%	65%	52%	61%	62%	59%	45%	59%	68%	67%	66%	59%	52%	59%	54%	65%	66%	62%	65%	57%	66%
75.3%	75%		78%							74%											74%
76.7%	70%																		80%		80%
81.7%	77%	89%	82%	90%	82%	80%	87%	73%	87%	90%	100%	78%	78%	77%	82%	87%	82%	76%	74%	80%	65%
56.5%	64%		64%	64%	50%					50%					50%	64%	40%		62%	64%	50%
73.8%	89%	74%	74%	74%	79%	75%	79%	71%	71%	75%	71%	69%	69%	69%	74%	75%	74%	74%		64%	75%
45.0%	45%
64.8%	66%	68%	66%	60%						66%	59%	66%	66%	75%	60%	61%
77.4%	76%		77%	77%	87%					77%					77%	79%	91%		68%		65%
81.0%	90%		90%	73%	80%					90%					85%	89%	82%		74%	77%	61%
70.7%	82%		82%	76%	70%					64%					73%	76%	56%		69%	68%	62%
62.9%	71%			68%						62%						69%	61%		52%	50%	70%
30.7%	34%	12%	41%	34%	69%	12%	34%	34%	28%	34%	34%	12%	28%	20%	47%	48%	23%	13%	20%	41%	26%
63.9%	61%	64%	70%	70%	68%	71%	67%	62%	67%	67%	67%	70%	61%	58%	58%	62%	64%	67%	67%	36%	64%
46.0%	46%
65.0%	65%
51.9%	48%	49%	56%	47%	56%	50%	56%	54%	49%	50%	56%	52%	56%	56%	47%	52%	56%	43%	43%	53%	61%
73.4%	73%		73%	71%						73%						68%	83%		70%	68%	82%
73.0%	73%
66.1%	59%		65%	67%	68%					67%					64%	67%	70%		74%	67%	59%
85.5%	89%									90%									74%		89%
80.0%	88%																				72%
73.3%	82%	91%	100%	82%	100%	65%	100%	68%	73%	59%	74%	43%	65%	65%	68%	82%	58%	56%	70%	79%	59%
70.4%	83%		75%	75%	76%					74%					73%	68%	64%		52%	68%	66%
70.5%	70%														71%
84.4%	80%		88%	77%	85%					85%					85%	86%	85%		92%	77%	88%
45.7%	48%		44%	54%	47%					47%					44%	46%	44%		41%	44%	44%
77.8%	79%	80%	79%	84%	74%	74%	79%	89%	88%	75%	89%	57%	76%	57%	80%	87%	89%	84%	71%	61%	82%
65.8%	66%	68%	71%	71%	47%	69%	68%	67%	63%	71%	63%	47%	74%	66%	68%	63%	63%	70%	54%	74%	78%
47.0%	47%
72.7%	64%																		80%		74%
62.5%	68%		68%	68%	68%					68%					68%	68%	45%		61%	53%	53%
71.5%	68%																				75%
91.9%	92%		96%	85%	92%					92%					96%	94%	92%		100%	76%	96%
71.0%	70%				84%														56%		74%
85.0%	84%	86%	90%	90%	71%	80%	90%	100%	84%	80%	81%	90%	80%	75%	80%	90%	92%	84%	90%	87%	82%
53.5%	58%		55%	57%	54%					57%					61%	55%	47%		55%	45%	45%

Model Similarity Dendrogram

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