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%3Apakistan C10 - Run b81b2c4...

Global Dialogues GD3 - Pakistan (10 context Qs)

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

TAGS:

Instruction Following & Prompt Adherence

Factual Accuracy & Hallucination

Reasoning

Efficiency & Succinctness

Best Models (Coverage)

1.Claude Sonnet 4
75.5%
2.Claude Sonnet 4.5
75.2%
3.GPT 5
74.8%
4.Claude 3.7 Sonnet
74.3%
5.Claude Haiku 4.5
74.2%

🔀 Least Similar Models

Claude Sonnet 4.5vsGPT OSS 20b

67.6% similarity

👯 Most Similar Models

Llama 4 MaverickvsGPT 4o

88.1% 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	4th 74.3%	5th 74.2%	1st 75.5%	2nd 75.2%	11th 72.3%	19th 67.1%	9th 73.4%	8th 73.6%	10th 73.0%	6th 74.0%	15th 68.7%	20th 66.7%	12th 71.6%	17th 67.7%	3rd 74.8%	7th 73.6%	14th 69.0%	21st 64.6%	13th 69.4%	18th 67.4%	16th 67.9%
73.6%	70%	76%	75%	69%	65%	78%	73%	70%	76%	75%	74%	76%	74%	76%	76%	79%	68%	70%	79%	71%	75%
73.0%	77%	73%	77%	80%	56%	84%	73%	81%	64%	79%	73%	75%	73%	75%	73%	61%	71%	63%	75%	58%	92%
67.7%	72%	67%	81%	60%	60%	73%	70%	72%	64%	75%	77%	63%	68%	68%	70%	69%	67%	60%	63%	60%	62%
78.4%	80%	76%	76%	76%	82%	71%	83%	80%	76%	83%	76%	79%	80%	71%	85%	83%	74%	83%	76%	76%	80%
89.3%	94%	93%	90%	98%	96%	90%	91%	87%	87%	85%	91%	87%	95%	91%	92%	92%	86%	88%	96%	79%	67%
62.0%	68%	67%	66%	72%	73%	59%	59%	59%	63%	66%	62%	59%	56%	56%	65%	62%	63%	62%	59%	53%	52%
87.1%	88%	80%	92%	92%	83%	85%	85%	90%	93%	90%	89%	79%	89%	93%		91%	84%	90%	81%	87%	82%
69.0%	83%	63%	72%	83%	66%	44%	71%	81%	81%	73%	57%	57%	68%	57%	85%	72%	77%	74%	60%	63%	63%
81.6%	73%	82%	77%	73%	82%	70%	79%	93%	87%	84%	80%	83%	83%	85%	84%	72%	89%	84%	83%	85%	85%
80.0%	82%	91%	89%	84%	79%	79%	79%	73%	79%	80%	75%	79%	79%	79%	84%	82%	79%	66%	79%	79%	85%
84.5%	87%	95%	91%	90%	74%	89%	89%	89%	93%	81%	86%	88%	77%	70%	89%	86%	84%	64%	83%	90%	80%
71.6%	81%	73%	75%	80%	72%	70%	67%	78%	80%	65%	66%	79%	56%	62%	78%	74%	61%	51%	74%	65%	96%
73.3%	61%	84%	76%	74%	78%	87%	72%	67%	67%	73%	80%	72%	71%	82%	63%	73%	80%	72%	82%	63%	63%
71.4%	73%	78%	80%	71%	77%	68%	74%	70%	76%	73%	80%	51%	78%	64%	75%	74%	63%	82%	79%	60%	53%
73.0%	76%	76%	81%	80%	72%	71%	78%	77%	75%	75%	72%	75%	75%	69%	73%	73%	67%	64%	65%	76%	64%
70.3%	77%	72%	79%	75%	81%	56%	70%	75%	67%	75%	66%	59%	75%	67%	77%	73%	70%	65%	64%	72%	62%
60.9%	71%	73%	75%	75%	72%	53%	74%	46%	64%	60%	62%	53%	50%	52%	68%	70%	52%	31%	66%	52%	60%
61.9%	76%	64%	76%	73%	72%	45%	57%	71%	61%	67%	44%	52%	69%	53%	71%	61%	65%	57%	56%	52%	58%
56.1%	58%	41%	54%	47%	66%	43%	47%	50%	56%	69%	41%	51%	75%	64%	72%	77%	54%	44%	41%	72%
80.1%	80%	91%	83%	86%	67%	71%	89%	78%	79%	85%	73%	67%	81%	73%	90%	92%	76%	77%	92%	68%	85%
83.7%	70%	92%	92%	78%	95%	66%	89%	85%	91%	92%	83%	81%	93%	80%	87%	83%	88%	72%	67%	90%	83%
77.3%	83%	86%	85%	83%	73%	64%	82%	85%	77%	83%	74%	76%	83%	75%	84%	77%	74%	66%	65%	77%	71%
61.7%	62%	68%	68%	60%	61%	63%	67%	69%	67%	61%	67%	36%	58%	63%	67%	59%	59%	57%	70%	57%	56%
80.8%	81%	83%	85%	81%	84%	81%	79%	82%	78%	82%	81%	75%	84%	84%	82%	85%	81%	81%	73%	74%
65.2%	67%	70%	70%	67%	71%	72%	70%	69%	65%	71%	65%	52%	65%	61%	64%	68%	60%	65%	67%	54%	56%
66.8%	62%	68%	68%	60%	69%	63%	68%	58%	62%	69%	74%	78%	67%	79%	62%	66%	67%	64%	72%	62%	64%
58.1%	59%	61%	66%	63%	61%	51%	55%	60%	57%	61%	52%	56%	61%	59%	61%	63%	54%	52%	57%	58%	54%
53.3%	66%	61%	61%	60%	70%	34%	64%	68%	72%	57%	38%	42%	49%	38%	54%	61%	55%	41%	52%	40%	37%
76.5%	85%	84%	86%	92%	83%	69%	81%	80%	81%	66%	59%	85%	57%	60%	100%	85%	63%	69%	73%	72%
63.9%	71%	64%	67%	80%	72%	55%	60%	65%	67%	73%	60%	58%	61%	53%	67%	66%	61%	56%	66%	65%	55%
62.2%	67%	66%	64%	73%	68%	57%	67%	74%	65%	67%	60%	47%	63%	55%	61%	65%	62%	49%	54%	58%	65%
67.1%	62%	68%	53%	75%	75%	62%	64%	62%	70%	59%	58%	71%	66%	66%	71%	65%	68%	75%	71%	75%	73%
50.3%	55%	50%	63%	44%	44%	55%	53%	55%	48%	57%	59%	47%	51%	51%	44%	51%	50%	44%	47%	44%	44%
74.5%	84%	73%	74%	82%	74%	68%	73%	77%	81%	75%	70%	63%	69%	63%	80%	81%	76%	82%	65%	73%	81%
70.6%	73%	76%	76%	74%	45%	73%	71%	76%	71%	64%	64%	65%	71%	74%	78%	74%	71%	74%	73%	74%	65%
71.2%	79%	72%	76%	80%	85%	62%	75%	72%	66%	82%	70%	65%	72%	65%	74%	76%	71%	59%	64%	61%	69%
76.1%	75%	78%	83%	92%	70%	74%	90%	93%	85%	79%	64%	79%	74%	61%	80%	71%	68%	53%	79%	75%	75%
81.3%	93%	86%	86%	86%	81%	90%	85%	88%	86%	89%	79%	72%	90%	70%	93%	90%	72%	69%	76%	74%	52%
54.1%	64%	58%	62%	64%	61%	48%	65%	65%	52%	66%	52%	48%	64%	38%	53%	51%	41%	31%	43%	45%	65%
92.9%	96%	94%	91%	87%	87%	100%	98%	100%	92%	97%	96%	90%	95%	95%	90%	97%	94%	87%	90%	87%	87%
62.2%	68%	68%	58%	65%	74%	54%	71%	65%	76%	63%	56%	41%	70%	64%	73%	69%	60%	35%	57%	55%	64%
80.3%	84%	81%	77%	80%	81%	75%	88%	81%	81%	91%	74%	84%	82%	81%	80%	78%	74%	78%	81%	77%	79%
65.5%	63%	70%	70%	70%	52%	65%	63%	48%	61%	63%	74%	72%	61%	70%	67%	67%	68%	72%	70%	70%	59%

Model Similarity Dendrogram

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