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 AgeGroup%3A46 55 C5 - Run 2cbae08...

Global Dialogues GD3 - 46-55 (5 context Qs)

DTEF: Predict response distributions for 46-55. 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.GPT 5
83.5%
2.GPT 5.1
83.4%
3.Claude 3.7 Sonnet
82.8%
4.GPT 4.1 Mini
82.6%
5.Claude Sonnet 4.5
82.5%

🔀 Least Similar Models

Claude Sonnet 4.5vsGPT OSS 20b

66.1% similarity

👯 Most Similar Models

Claude 3.7 SonnetvsLlama 4 Maverick

88.9% 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 4.1 Fast
Score	3rd 82.8%	6th 82.2%	8th 81.7%	5th 82.5%	12th 80.0%	20th 76.0%	10th 80.6%	7th 81.8%	11th 80.3%	14th 79.9%	4th 82.6%	19th 77.1%	9th 81.3%	18th 77.7%	1st 83.5%	2nd 83.4%	16th 78.6%	21st 74.2%	15th 79.0%	17th 78.1%	13th 79.9%
83.1%	88%	86%	72%	77%	81%	66%	84%	90%	85%	80%	86%	84%	92%	80%	81%	88%	86%	84%	85%	82%	89%
89.7%	93%	89%	95%	93%	66%	97%	94%	93%	81%	97%	94%	87%	93%	85%	95%	90%	88%	80%	90%	94%	90%
69.4%	67%	67%	78%	70%	60%	75%	67%	73%	63%	74%	71%	75%	66%	67%	70%	69%	79%	64%	62%	70%	70%
87.5%	91%	84%	89%	93%	76%	86%	87%	87%	84%	92%	89%	86%	90%	75%	89%	90%	92%	89%	80%	94%	94%
89.3%	89%	95%	94%	96%	86%	91%	95%	85%	89%	97%	89%	82%	89%	91%	96%	88%	94%	78%	98%	80%	74%
92.1%	98%	97%	94%	97%	89%	98%	92%	89%	89%	92%	91%	78%	94%	86%	93%	97%	86%	90%	96%	92%	96%
83.9%	80%	83%	82%	93%	76%	82%	83%	92%	84%	79%	89%	92%	82%	79%	83%	90%	84%	80%	85%	80%	84%
74.8%	72%	71%	71%	75%	78%	50%	73%	84%	83%	78%	77%	78%	75%	72%	78%	83%	73%	66%	64%	80%	89%
84.8%	88%	84%	80%	84%	83%	74%	82%	82%	80%	89%	92%	92%	85%	85%	85%	90%	85%	83%	86%	90%	82%
88.9%	95%	85%	94%	96%	88%	91%	92%	86%	88%	95%	91%	95%	97%	87%	88%	95%	76%	87%	91%	63%	86%
73.1%	85%	75%	85%	85%	56%	83%	71%	82%	70%	66%	72%	73%	67%	73%	72%	68%	59%	56%	82%	70%	85%
78.9%	84%	77%	80%	83%	81%	79%	77%	86%	81%	77%	80%	76%	79%	80%	79%	74%	78%	71%	79%	77%	78%
88.0%	89%	91%	90%	92%	87%	90%	86%	77%	88%	92%	90%	79%	90%	89%	91%	91%	84%	89%	87%	87%	90%
82.0%	83%	81%	81%	82%	88%	82%	82%	79%	85%	79%	89%	78%	79%	84%	82%	88%	74%	83%	90%	81%	73%
83.3%	82%	91%	84%	87%	84%	81%	90%	83%	86%	81%	87%	83%	89%	84%	83%	81%	84%	59%	83%	84%	84%
82.0%	85%	84%	86%	90%	92%	60%	78%	89%	80%	78%	85%	70%	86%	73%	91%	89%	87%	77%	78%	78%	86%
67.5%	70%	83%	87%	72%	76%	75%	67%	57%	63%	58%	69%	57%	61%	62%	75%	73%	56%	49%	67%	68%	72%
58.8%	64%	53%	55%	64%	57%	42%	46%	67%	59%	52%	53%	47%	61%	51%	69%	66%	66%	65%	60%	69%	69%
52.6%	62%	39%	63%	58%	60%	42%	61%	52%	54%	51%	50%	40%	49%	47%	79%	73%	46%	47%	37%	47%	47%
83.2%	74%	75%	74%	73%	94%	91%	83%	86%	93%	82%	90%	86%	90%	82%	79%	80%	82%	71%	86%	94%	82%
81.8%	84%	83%	81%	81%	84%	66%	89%	87%	84%	80%	84%	83%	82%	80%	90%	93%	87%	72%	75%	84%	69%
82.1%	86%	85%	78%	77%	89%	79%	89%	87%	85%	81%	83%	81%	86%	84%	87%	88%	74%	65%	76%	84%	80%
76.5%	74%	75%	79%	79%	86%	73%	83%	83%	76%	77%	82%	67%	76%	77%	76%	73%	74%	74%	74%		72%
84.1%	91%	93%	92%	91%	85%	81%	83%	78%	84%	92%	90%	71%	90%	85%	84%	87%	80%	60%	83%	80%	86%
77.8%	81%	78%	78%	77%	84%	79%	78%	74%	83%	71%	83%	69%	77%	81%	77%	79%	68%	87%	92%	71%	67%
82.9%	88%	83%	85%	81%	85%	75%	82%	78%	82%	90%	86%	83%	88%	78%	80%	80%	85%	86%	84%	91%	70%
72.6%	78%	68%	69%	73%	73%	50%	67%	81%	75%	70%	61%	75%	70%	52%	83%	82%	86%	70%	73%	84%	84%
81.6%	88%	91%	82%	82%	90%	62%	83%	91%	86%	76%	82%	78%	79%	68%	91%	84%	78%	80%	77%	84%	82%
75.1%	86%	88%	85%	85%	85%	72%	79%	71%	70%	69%	71%	67%	68%	68%	77%	82%	70%	71%	69%	72%	72%
75.9%	75%	81%	75%	86%	91%	64%	64%	80%	79%	73%	75%	69%	74%	76%	77%	70%	80%	75%	73%	80%	77%
85.5%	90%	91%	86%	81%	91%	88%	84%	80%	87%	83%	91%	79%	90%	90%	91%	91%	84%	57%	89%	89%	83%
80.4%	85%	84%	72%	81%	92%	76%	84%	78%	88%	77%	80%	76%	85%	84%	81%	80%	71%	83%	90%	61%	81%
65.8%	63%	63%	75%	66%	56%	71%	63%	69%	60%	70%	67%	71%	62%	63%	66%	67%	76%	64%	58%	66%	66%
78.5%	84%	84%	80%	85%	87%	70%	83%	86%	84%	84%	86%	70%	75%	68%	82%	80%	63%	81%	70%	70%	76%
77.6%	87%	84%	77%	77%	60%	90%	79%	78%	72%	76%	75%	87%	87%	73%	70%	76%	81%	85%	69%	72%	75%
84.6%	94%	91%	94%	90%	69%	69%	91%	97%	89%	95%	88%	81%	87%	71%	83%	89%	77%	75%	79%	78%	89%
87.9%	89%	94%	88%	89%	91%	89%	94%	82%	88%	91%	92%	81%	93%	92%	94%	93%	83%	67%	85%	84%	86%
87.1%	84%	99%	92%	91%	87%	95%	85%	95%	85%	85%	90%	79%	83%	87%	96%	88%	88%	85%	94%	59%	82%
81.5%	87%	79%	87%	84%	88%	70%	87%	88%	83%	74%	86%	83%	83%	75%	90%	86%	77%	75%	77%	82%	71%
94.5%	94%	95%	93%	98%	87%	97%	95%	99%	91%	98%	99%	97%	94%	95%	98%	99%	92%	96%	89%	81%	98%
84.0%	89%	90%	92%	94%	78%	65%	92%	93%	88%	94%	89%	71%	90%	84%	95%	94%	84%	41%	69%	81%	90%
83.5%	86%	83%	76%	80%	83%	71%	76%	85%	80%	76%	91%	79%	88%	97%	85%	86%	88%	89%	89%		83%
72.7%	59%	83%	65%	61%	62%	80%	64%	57%	69%	66%	87%	80%	74%	80%	78%	75%	74%	85%	79%	71%	78%

Model Similarity Dendrogram

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