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%3Aindia C5 - Run 2037ccc...

Global Dialogues GD3 - India (5 context Qs)

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

TAGS:

Instruction Following & Prompt Adherence

Reasoning

Factual Accuracy & Hallucination

Efficiency & Succinctness

Best Models (Coverage)

1.Claude Sonnet 4
83.5%
2.Claude 3.7 Sonnet
82.9%
3.Claude Sonnet 4.5
82.6%
4.GPT 4.1
81.9%
5.Claude Haiku 4.5
81.9%

🔀 Least Similar Models

Claude Sonnet 4.5vsGPT OSS 20b

65.0% similarity

👯 Most Similar Models

GPT 4o MinivsGPT 4o

87.5% 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	2nd 82.9%	5th 81.9%	1st 83.5%	3rd 82.6%	7th 80.9%	21st 71.0%	11th 79.6%	10th 79.7%	11th 79.6%	4th 81.9%	13th 79.4%	15th 76.3%	8th 80.9%	16th 75.6%	9th 80.7%	6th 81.7%	17th 75.2%	20th 73.5%	19th 73.9%	18th 74.9%	14th 77.8%
79.7%	86%	87%	90%	92%	81%	68%	76%	71%	72%	82%	78%	79%	85%	74%	83%	85%	82%	71%	80%	77%	75%
77.4%	74%	78%	76%	80%	63%	92%	70%	83%	66%	71%	89%	88%	85%	85%	78%	68%	74%	58%	87%	84%	77%
71.8%	81%	68%	78%	81%	73%	71%	66%	78%	64%	84%	75%	66%	69%	69%	76%	71%	70%	73%	66%	64%	64%
91.7%	92%	91%	95%	96%	82%	89%	94%	92%	94%	95%	93%	94%	93%	89%	91%	92%	90%	88%	87%	93%	95%
89.3%	89%	94%	89%	88%	94%	82%	88%	94%	95%	84%	90%	92%	94%	91%	85%	94%	92%	88%	80%	96%	76%
84.6%	87%	90%	91%	91%	98%	87%	84%	87%	90%	86%	88%	67%	86%	69%	86%	82%	78%	84%	85%	71%	89%
78.1%	87%	84%	86%	85%	88%	64%	79%	84%	86%	77%	79%	58%	74%	79%	77%	90%	87%	70%	75%	63%	69%
78.2%	92%	75%	75%	86%	78%	42%	74%	94%	86%	92%	68%	73%	77%	66%	95%	92%	73%	76%	63%	83%	82%
80.2%	75%	85%	79%	79%	94%	62%	79%	88%	84%	78%	88%	75%	87%	85%	72%	80%	83%	88%	75%	74%	74%
89.4%	92%	93%	91%	94%	87%	91%	93%	88%	88%	96%	92%	97%	92%	87%	92%	90%	77%	79%	90%	79%	90%
82.6%	94%	92%	89%	90%	78%	86%	87%	77%	86%	78%	88%	85%	78%	78%	82%	85%	79%	63%	73%	90%	77%
88.8%	87%	90%	92%	91%	90%	90%	93%	85%	86%	90%	87%	94%	92%	87%	92%	94%	85%	77%	88%	87%	87%
82.6%	71%	88%	84%	76%	83%	80%	83%	85%	90%	85%	88%	78%	87%	85%	78%	75%	83%	89%	77%	85%	84%
78.8%	78%	92%	85%	78%	70%	78%	90%	70%	83%	76%	91%	73%	85%	96%	70%	83%	68%	65%	93%	65%	66%
84.0%	94%	87%	94%	95%	89%	66%	93%	92%	91%	93%	83%	90%	88%	79%	72%	88%	71%	72%	73%	74%	81%
79.3%	89%	90%	89%	86%	76%	59%	71%	81%	73%	88%	77%	75%	94%	68%	88%	93%	79%	74%	75%	65%	76%
61.9%	73%	69%	73%	72%	66%	54%	65%	48%	63%	68%	61%	44%	49%	53%	75%	72%	47%	43%	66%	65%	73%
69.8%	92%	67%	82%	89%	87%	50%	59%	68%	58%	82%	61%	63%	72%	51%	83%	79%	67%	71%	62%	61%	61%
51.4%	47%	44%	52%	48%	66%	46%	58%	51%	52%	48%	49%	44%	47%	47%	83%	62%	54%	43%	38%	47%	53%
84.0%	84%	78%	88%	76%	80%	89%	94%	67%	78%	86%	91%	81%	93%	92%	77%	83%	87%	80%	87%	89%	83%
83.2%	86%	92%	92%	86%	63%	67%	84%	85%	89%	88%	85%	89%	87%	79%	89%	84%	77%	86%	66%	78%	96%
81.3%	82%	79%	84%	87%	88%	71%	85%	79%	83%	84%	79%	89%	84%	81%	80%	84%	73%	82%	68%	82%	84%
73.8%	73%	74%	73%	73%	81%	84%	75%	82%	68%	70%	73%	63%	76%	86%	74%	69%	75%	68%	65%	72%	75%
85.6%	79%	86%	83%	81%	90%	83%	89%	89%	89%	84%	93%	74%	92%	92%	84%	85%	88%	80%	81%	92%	83%
81.5%	79%	88%	84%	75%	86%	80%	85%	72%	82%	72%	89%	82%	82%	93%	79%	84%	72%	82%	87%	77%	82%
73.6%	78%	76%	78%	70%	82%	69%	72%	67%	73%	70%	79%	80%	78%	73%	71%	73%	72%	77%	80%	75%	53%
65.7%	80%	71%	80%	80%	68%	38%	55%	67%	64%	78%	52%	64%	70%	48%	76%	69%	70%	66%	62%	52%	70%
76.5%	91%	79%	88%	83%	90%	52%	78%	91%	85%	93%	67%	82%	72%	58%	92%	88%	58%	71%	55%	54%	79%
73.7%	85%	92%	84%	84%	93%	67%	78%	69%	70%	65%	68%	63%	66%	64%	74%	76%	71%	69%	65%	69%	76%
77.4%	95%	88%	90%	92%	78%	61%	61%	85%	85%	93%	76%	70%	76%	63%	82%	76%	71%	87%	74%	60%	62%
83.9%	92%	89%	93%	96%	93%	67%	92%	87%	89%	92%	78%	89%	90%	76%	67%	68%	86%	68%	71%	85%	94%
78.4%	70%	81%	71%	68%	76%	77%	83%	71%	84%	65%	73%	83%	80%	80%	76%	82%	84%	89%	83%	85%	85%
65.7%	76%	59%	73%	76%	68%	67%	62%	73%	60%	78%	70%	62%	64%	64%	69%	63%	60%	56%	61%	59%	59%
79.2%	76%	81%	82%	72%	74%	81%	88%	83%	84%	77%	84%	69%	85%	71%	81%	76%	86%	72%	74%	76%	92%
81.8%	74%	84%	87%	77%	82%	86%	78%	83%	81%	93%	81%	77%	91%	80%	73%	90%	79%	69%	69%	97%	86%
83.8%	92%	91%	96%	96%	65%	68%	92%	97%	90%	86%	83%	77%	90%	75%	69%	90%	77%	77%	79%	83%	86%
80.6%	85%	85%	85%	90%	81%	64%	83%	81%	83%	85%	76%	87%	81%	75%	91%	92%	73%	74%	70%	59%	93%
87.7%	92%	92%	98%	91%	85%	90%	89%	96%	90%	93%	93%	72%	94%	76%	97%	95%	72%	73%	92%	75%	86%
77.3%	92%	80%	83%	87%	81%	59%	82%	88%	76%	92%	74%	79%	80%	65%	94%	86%	59%	55%	59%	82%	71%
94.2%	94%	96%	94%	91%	99%	100%	94%	94%	92%	88%	97%	94%	96%	97%	95%	95%	98%	88%	94%	91%	91%
76.1%	79%	75%	71%	75%	89%	58%	79%	81%	87%	91%	82%	65%	72%	72%	85%	87%	73%	67%	57%	74%	80%
74.2%	76%	75%	74%	77%	77%	62%	75%	79%	77%	79%	67%	83%	74%	80%	71%	73%	70%	72%	79%	69%	70%
67.0%	73%	67%	70%	70%	67%	57%	69%	47%	58%	68%	79%	71%	70%	71%	67%	70%	62%	79%	67%	62%	62%

Model Similarity Dendrogram

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