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 AiConcern%3Amore Excited Than Concerned C5 - Run b259408...

Global Dialogues GD3 - More excited than concerned (5 context Qs)

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

TAGS:

Instruction Following & Prompt Adherence

Factual Accuracy & Hallucination

Reasoning

AI Safety & Robustness

Social interaction

Best Models (Coverage)

1.Claude 3.7 Sonnet
85.1%
2.Claude Sonnet 4
84.7%
3.Claude Haiku 4.5
84.7%
4.GPT 5.1
84.3%
5.Claude Sonnet 4.5
84.2%

🔀 Least Similar Models

Mistral Large 2411vsGPT 4.1 Nano

65.3% similarity

👯 Most Similar Models

Claude Haiku 4.5vsClaude Sonnet 4

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
Score	1st 85.1%	3rd 84.7%	2nd 84.7%	5th 84.2%	17th 73.9%	19th 73.1%	10th 82.2%	11th 80.5%	8th 82.6%	9th 82.2%	7th 84.0%	21st 47.1%	13th 77.3%	14th 76.7%	6th 84.1%	4th 84.3%	20th 71.0%	18th 73.7%	12th 77.9%	15th 75.4%	16th 75.1%
82.8%	91%	93%	90%	91%	66%	77%	87%	93%	90%	96%	86%	21%	79%	67%	93%	86%	78%	82%	90%	88%	94%
74.8%	79%	74%	80%	94%	64%	77%	94%	78%	68%	78%	85%	16%	72%	72%	90%	85%	77%	76%	71%	57%	84%
84.6%	82%	85%	85%	85%	96%	10%	74%	91%	81%	90%	98%	79%	79%	94%	88%	77%	98%	92%	98%	98%	96%
91.0%	96%	86%	91%	90%	88%	94%	93%	90%	96%	92%	97%	88%	97%	87%	93%	94%	87%	83%	87%	92%	90%
86.1%	91%	95%	81%	78%	85%	82%	83%	95%	91%	94%	97%	72%	76%	90%	87%	86%	92%	72%	75%	92%	94%
89.0%	90%	95%	92%	92%	84%	81%	96%	91%	97%	92%	95%	97%	92%	88%	83%	87%	93%	91%	84%	71%	77%
80.7%	86%	95%	88%	90%	58%	80%	86%	97%	93%	84%	94%	57%	72%	68%	90%	96%	84%	89%	50%	64%	74%
75.6%	90%	82%	78%	91%	79%	77%	83%	85%	79%	94%	89%	47%	87%	65%	94%	94%	28%	30%	81%	73%	62%
78.0%	83%	82%	86%	85%	94%	65%	81%	80%	81%	80%	80%	51%	80%	69%	75%	84%	77%	69%	76%	84%	76%
86.8%	96%	93%	85%	93%	93%	92%	85%	79%	83%	86%	88%	91%	84%	88%	87%	86%	75%	73%	87%	81%	97%
74.7%	85%	84%	87%	85%	55%	88%	69%	89%	78%	70%	84%	54%	54%	82%	69%	72%	89%	63%	85%	86%	41%
89.3%	86%	91%	91%	95%	93%	89%	84%	90%	92%	96%	85%	93%	90%	87%	91%	88%	82%	83%	90%	90%	90%
78.9%	82%	83%	88%	80%	92%	86%	81%	74%	89%	77%	90%	15%	79%	75%	91%	92%	84%	79%	80%	57%	82%
64.5%	85%	76%	84%	78%	37%	66%	84%	74%	67%	77%	64%	32%	79%	44%	80%	80%	49%	69%	34%	52%	44%
77.7%	88%	87%	94%	86%	87%	93%	88%	81%	86%	84%	88%	9%	89%	84%	78%	89%	58%	80%	73%	68%	42%
76.5%	90%	88%	85%	90%	52%	82%	88%	87%	86%	91%	86%	23%	79%	73%	81%	84%	67%	70%	80%	67%	58%
81.3%	83%	91%	94%	75%	86%	89%	68%	66%	73%	74%	81%	83%	73%	86%	84%	82%	81%	77%	91%	89%	82%
75.8%	82%	79%	75%	81%	78%	44%	64%	80%	76%	83%	89%	28%	90%	86%	87%	89%	74%	84%	78%	78%	67%
55.5%	59%	48%	70%	61%	80%	54%	62%	55%	60%	56%	55%	40%	56%	48%	69%	65%	48%	38%	45%	50%	47%
80.8%	87%	96%	85%	92%	70%	85%	86%	81%	86%	82%	87%	21%	89%	78%	92%	83%	69%	79%	88%	67%	94%
75.3%	82%	89%	85%	86%	66%	84%	93%	73%	84%	86%	83%	34%	78%	75%	78%	84%	48%	63%	81%	82%	47%
78.9%	80%	84%	87%	86%	56%	92%	93%	80%	92%	84%	91%	23%	74%	91%	94%	88%	57%	79%	69%	63%	94%
76.9%	70%	71%	71%	76%	93%	89%	88%	81%	75%	79%	87%	64%	75%	91%	76%	79%	76%	64%	71%	70%	69%
72.7%	86%	91%	86%	83%	79%	86%	73%	83%	85%	87%	85%	6%	70%	57%	92%	97%	58%	59%	74%	59%	31%
69.6%	79%	87%	84%	78%	60%	77%	85%	78%	83%	73%	71%	22%	71%	53%	69%	84%	43%	73%	47%	80%	65%
68.7%	83%	80%	83%	79%	59%	63%	76%	76%	82%	74%	72%	18%	69%	66%	70%	74%	57%	62%	80%	46%	73%
76.4%	86%	83%	82%	84%	73%	47%	69%	87%	78%	88%	71%	45%	80%	75%	86%	89%	62%	82%	80%	71%	86%
85.5%	87%	97%	90%	87%	54%	85%	95%	91%	89%	94%	90%	46%	88%	91%	89%	90%	90%	82%	86%	91%	83%
77.9%	93%	90%	94%	94%	90%	70%	84%	71%	70%	66%	73%	78%	72%	65%	76%	72%	67%	72%	74%	70%	95%
79.0%	75%	83%	78%	84%	69%	74%	81%	77%	78%	85%	88%	72%	82%	64%	85%	76%	80%	72%	86%	85%	84%
74.7%	89%	75%	81%	82%	83%	91%	82%	77%	87%	75%	85%	8%	68%	79%	88%	91%	68%	59%	87%	74%	39%
62.4%	70%	70%	70%	70%	60%	10%	81%	64%	74%	65%	55%	41%	41%	51%	83%	78%	55%	64%	54%	58%	96%
83.2%	80%	82%	82%	82%	93%	10%	72%	89%	79%	88%	99%	81%	81%	96%	76%	74%	99%	89%	100%	96%	99%
82.6%	82%	93%	86%	88%	58%	78%	85%	82%	93%	84%	93%	37%	90%	79%	90%	85%	85%	86%	90%		89%
77.9%	94%	82%	78%	77%	78%	71%	65%	70%	87%	78%	85%	70%	87%	87%	62%	64%	70%	83%	71%	88%	88%
73.9%	75%	80%	82%	82%	40%	84%	83%	86%	79%	79%	73%	32%	79%	74%	65%	85%	74%	60%	82%	83%	75%
76.0%	89%	76%	84%	84%	83%	89%	85%	65%	78%	74%	81%	8%	81%	78%	78%	78%	61%	73%	88%	80%	83%
81.9%	94%	96%	90%	87%	76%	78%	80%	93%	99%	90%	86%	62%	84%	83%	91%	85%	60%	70%	82%	86%	47%
79.7%	92%	93%	95%	91%	67%	91%	88%	82%	86%	85%	89%	28%	79%	89%	89%	90%	68%	63%	89%	72%	48%
78.9%	91%	91%	91%	91%	80%	10%	98%	84%	94%	85%	74%	58%	58%	70%	95%	99%	74%	84%	74%	78%	77%
84.3%	95%	87%	91%	94%	54%	86%	90%	93%	83%	90%	95%	39%	92%	89%	97%	85%	64%	92%	88%	83%	84%
72.5%	100%	66%	72%	66%	80%	76%	70%	52%	67%	60%	72%	59%	55%	66%	100%	100%	62%	67%	69%	63%	100%
84.9%	77%	91%	91%	79%	88%	93%	81%	73%	79%	89%	95%	76%	76%	97%	85%	78%	83%	93%	86%	85%	87%

Model Similarity Dendrogram

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