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 Religion%3Ajudaism C5 - Run 89700a2...

Global Dialogues GD3 - Judaism (5 context Qs)

DTEF: Predict response distributions for Judaism. 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
78.0%
2.Claude Sonnet 4.5
77.4%
3.Llama 4 Maverick
76.0%
4.Claude 3.7 Sonnet
75.9%
5.GPT 4.1 Mini
75.9%

🔀 Least Similar Models

Claude Sonnet 4.5vsGPT OSS 20b

61.2% similarity

👯 Most Similar Models

Llama 4 MaverickvsMistral Medium 3

91.7% 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 4.1 Fast
Score	4th 75.9%	6th 75.5%	1st 78.0%	2nd 77.4%	15th 71.5%	12th 73.9%	3rd 76.0%	14th 73.4%	13th 73.9%	8th 74.9%	5th 75.9%	19th 68.5%	7th 75.3%	10th 74.5%	18th 69.0%	9th 74.6%	20th 67.3%	21st 65.7%	11th 74.3%	17th 70.3%	15th 71.5%
76.0%	76%	75%	80%	79%	72%	62%	84%	81%	81%	98%	77%	81%	81%	82%	82%	66%	44%	74%	77%	68%	75%
63.7%	71%	54%	71%	79%	34%	71%	60%	61%	49%	59%	75%	62%	67%	62%	77%	74%	63%	52%	70%	71%	56%
67.2%	73%	65%	76%	73%	55%	70%	61%	67%	61%	73%	77%	57%	73%	70%	73%	61%	73%	59%	57%	65%	73%
77.9%	85%	73%	79%	79%	86%	66%	83%	83%	85%	78%	80%	74%	74%	84%	77%	74%	63%	78%	78%	79%	78%
78.3%	86%	83%	84%	86%	81%	91%	86%	70%	76%	67%	77%	69%	76%	80%	93%	85%	63%	59%	86%	81%	65%
85.3%	82%	82%	74%	85%	89%	78%	88%	92%	93%	88%	88%	80%	98%	92%	82%	82%	92%	79%	87%	80%	80%
84.1%	86%	82%	91%	93%	91%	79%	86%	78%	97%	78%	88%	78%	70%	83%	89%	77%	91%	88%	70%	97%	75%
76.8%	82%	79%	73%	74%	81%	67%	82%	81%	78%	77%	81%	78%	82%	82%	67%	81%	82%	54%	75%	81%	75%
68.2%	59%	77%	74%	74%	93%	75%	79%	75%	80%	71%	67%	71%	71%	57%	52%	60%	54%	80%	69%	47%	47%
73.2%	78%	81%	74%	74%	74%	85%	70%	70%	74%	74%	74%	74%	70%	78%	78%	69%	54%	65%	84%	68%	70%
63.9%	78%	64%	76%	76%	64%	67%	76%	67%	61%	56%	65%	61%	59%	59%	44%	58%	64%	29%	77%	58%	83%
80.9%	72%	77%	95%	75%	90%	82%	88%	74%	89%	83%	86%	78%	88%	85%	87%	78%	81%	60%	74%	88%	69%
71.3%	67%	76%	75%	79%	78%	82%	68%	64%	68%	69%	74%	42%	73%	76%	83%	75%	58%	65%	78%	69%	78%
69.6%	76%	71%	75%	67%	68%	81%	73%	71%	65%	75%	71%	76%	77%	66%	71%	61%	61%	56%	73%	62%	65%
77.0%	85%	76%	77%	77%	77%	77%	80%	70%	77%	76%	84%	67%	89%	81%	82%	82%	65%	58%	77%	80%	81%
72.5%	76%	77%	79%	74%	47%	64%	76%	75%	71%	81%	80%	80%	80%	75%	71%	74%	72%	67%	75%	59%	69%
78.9%	79%	75%	70%	82%	78%	79%	78%	83%	76%	84%	83%	82%	76%	80%	80%	78%	74%	70%	83%	83%	83%
72.5%	81%	79%	72%	80%	83%	50%	66%	73%	73%	79%	67%	66%	77%	67%	81%	76%	74%	69%	67%	76%	67%
50.7%	57%	40%	59%	55%	82%	34%	52%	45%	46%	50%	55%	38%	50%	46%	57%	65%	46%	50%	37%	55%	46%
69.1%	77%	73%	76%	77%	72%	73%	75%	61%	60%	72%	69%	59%	68%	72%	67%	78%	65%	68%	74%	51%	65%
73.8%	68%	74%	85%	76%	77%	90%	78%	58%	78%	70%	84%	67%	80%	89%	58%	76%	55%	60%	89%	71%	67%
72.7%	70%	74%	79%	78%	72%	82%	69%	72%	69%	79%	79%	59%	76%	77%	71%	79%	70%	64%	80%	51%	77%
67.0%	55%	67%	74%	68%	84%	85%	74%	80%	63%	67%	65%	51%	67%	83%	71%	67%	60%	59%	65%	51%	51%
68.4%	83%	73%	75%	77%	67%	71%	56%	56%	67%	75%	71%	34%	71%	61%	73%	79%	41%	65%	82%	67%	93%
69.0%	74%	76%	71%	64%	82%	74%	71%	67%	67%	57%	76%	62%	67%	68%	69%	66%	57%	72%	70%	71%	68%
71.7%	66%	77%	83%	77%	69%	82%	70%	70%	74%	72%	82%	67%	64%	79%	64%	64%	56%	64%	92%	74%	60%
71.9%	72%	75%	80%	77%	76%	45%	83%	83%	72%	79%	56%	82%	67%	67%	82%	75%	58%	59%	66%	69%	87%
74.6%	76%	79%	79%	79%	62%	79%	84%	69%	69%	73%	84%	76%	84%	83%	71%	60%	82%	68%	74%	85%	51%
63.3%	69%	79%	73%	77%	71%	62%	67%	65%	58%	55%	61%	51%	58%	55%	61%	69%	53%	54%	63%		64%
71.5%	81%	80%	74%	74%	65%	69%	63%	75%	77%	69%	75%	56%	70%	67%	73%	75%	73%	73%	72%	67%	73%
67.7%	75%	71%	71%	73%	50%	73%	70%	66%	67%	74%	71%	62%	69%	68%	72%	73%	55%	49%	67%	75%	71%
88.9%	83%	91%	80%	83%	98%	86%	96%	89%	96%	83%	79%	100%	83%	86%	83%	95%	83%	85%	100%	96%	92%
64.0%	65%	65%	76%	73%	55%	70%	61%	67%	61%	73%	77%	57%	73%	70%	0%	61%	65%	71%	57%	73%	73%
81.7%	88%	87%	85%	84%	60%	81%	86%	88%	88%	74%	87%	81%	80%	90%	80%	91%	78%	85%	76%	77%	70%
76.2%	81%	78%	88%	86%	43%	76%	76%	76%	88%	86%	88%	57%	85%	69%	86%	84%	82%	70%	64%	66%	72%
73.2%	75%	81%	84%	82%	37%	85%	80%	79%	63%	73%	75%	79%	79%	80%	54%	71%	79%	68%	77%	64%	73%
57.7%	64%	64%	65%	64%	62%	57%	64%	58%	58%	65%	62%	56%	62%	61%	0%	61%	54%	60%	59%	61%	54%
84.7%	93%	84%	92%	92%	79%	94%	93%	94%	92%	94%	89%	76%	88%	82%	83%	89%	63%	66%	80%	75%	80%
76.9%	79%	84%	84%	79%	56%	78%	82%	76%	81%	86%	79%	78%	82%	73%	80%	79%	59%	74%	64%	77%	84%
85.2%	83%	91%	80%	83%	98%	86%	96%	89%	96%	83%	79%	100%	83%	86%	0%	95%	92%	98%	100%	83%	88%
72.0%	81%	80%	81%	84%	68%	65%	79%	79%	79%	84%	66%	63%	84%	71%	84%	84%	84%	23%	67%	23%	82%
74.4%	72%	80%	72%	76%	59%	65%	76%	76%	72%		80%	78%	79%	81%	77%	75%	74%	77%	81%	66%	72%
83.4%	85%	78%	92%	84%	90%	90%	81%	82%	81%	86%	79%	81%	86%	80%	81%	87%	81%	82%	81%	91%	73%

Model Similarity Dendrogram

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