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 Environment%3Asuburban C10 - Run c6daff9...

Global Dialogues GD3 - Suburban (10 context Qs)

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

TAGS:

Instruction Following & Prompt Adherence

Reasoning

Efficiency & Succinctness

Factual Accuracy & Hallucination

Best Models (Coverage)

1.Claude Haiku 4.5
85.3%
2.Claude 3.7 Sonnet
84.7%
3.GPT 5.1
84.1%
4.GPT 5
84.1%
5.GPT 4.1
84.0%

🔀 Least Similar Models

Claude Sonnet 4.5vsGPT OSS 20b

70.8% similarity

👯 Most Similar Models

Claude Sonnet 4.5vsClaude Sonnet 4

90.6% 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 84.7%	1st 85.3%	6th 83.8%	18th 77.5%	10th 81.6%	16th 78.8%	11th 81.0%	9th 81.6%	12th 80.3%	5th 84.0%	8th 82.3%	16th 78.8%	7th 82.6%	14th 79.6%	3rd 84.1%	3rd 84.1%	20th 76.9%	21st 75.0%	13th 79.8%	19th 77.1%	15th 79.5%
85.5%	91%	89%	86%	87%	89%	73%	85%	82%	81%	87%	85%	83%	90%	82%	91%	89%	93%	81%	91%	82%	79%
86.5%	83%	85%	94%	94%	82%	88%	82%	83%	75%	89%	95%	90%	91%	91%	90%	97%	80%	69%	84%	85%	90%
66.9%	74%	68%	74%	0%	61%	77%	65%	72%	66%	77%	77%	80%	71%	74%	61%	71%	66%	68%	63%	69%	70%
89.3%	93%	75%	91%	91%	93%	85%	95%	95%	92%	93%	94%	92%	93%	79%	93%	81%	84%	85%	89%	88%	94%
88.0%	89%	90%	85%	90%	85%	83%	89%	91%	89%	98%	99%	93%	96%	93%	93%	93%	76%	78%	80%		71%
88.0%	89%	91%	84%	91%	92%	95%	89%	92%	90%	83%	83%	85%	82%	88%	80%	84%	90%	94%	84%	85%	97%
90.3%	88%	91%	91%	91%	76%	95%	86%	93%	92%	91%	90%	92%	85%	92%	94%	89%	97%	91%	87%		95%
80.1%	85%	85%	76%	77%	76%	55%	82%	88%	84%	83%	75%	72%	80%	72%	92%	93%	89%	84%	71%	90%	74%
79.7%	89%	77%	73%	77%	82%	64%	75%	74%	74%	91%	82%	87%	83%	78%	84%	87%	89%	89%	72%	75%	72%
88.9%	91%	86%	92%	91%	92%	94%	86%	86%	89%	92%	80%	92%	87%	92%	91%	87%	84%	76%	92%	92%	94%
83.5%	91%	90%	90%	91%	90%	90%	89%	85%	90%	78%	85%	92%	80%	82%	81%	78%	65%	59%	84%	80%	84%
86.8%	91%	89%	93%	93%	91%	88%	82%	92%	90%	74%	89%	90%	77%	84%	84%	89%	81%	84%	91%	85%	86%
82.4%	80%	93%	83%	92%	89%	94%	77%	76%	82%	85%	86%	74%	85%	91%	77%	87%	72%	83%	81%	72%	72%
87.2%	88%	96%	93%	96%	87%	79%	95%	83%	91%	93%	83%	68%	93%	66%	93%	86%	87%	83%	92%	92%	88%
86.1%	90%	88%	82%	82%	92%	88%	89%	85%	82%	93%	89%	80%	91%	95%	92%	95%	69%	64%	84%	88%	90%
82.4%	89%	84%	93%	90%	85%	76%	74%	87%	81%	85%	74%	80%	86%	79%	85%	88%	85%	82%	81%	78%	69%
62.5%	69%	86%	70%	70%	55%	64%	65%	46%	62%	57%	67%	71%	50%	60%	69%	67%	46%	56%	65%	51%	67%
72.1%	77%	73%	73%	80%	79%	56%	62%	81%	64%	66%	64%	70%	76%	63%	75%	79%	82%	78%	71%	71%	74%
69.4%	72%	64%	83%	70%	69%	51%	67%	76%	72%	90%	63%	55%	75%	64%	89%	91%	60%	60%	54%	65%	67%
84.9%	93%	98%	89%	92%	93%	76%	91%	82%	75%	95%	79%	76%	87%	77%	96%	87%	75%	84%	77%	77%	83%
87.4%	90%	92%	87%	85%	95%	78%	92%	90%	93%	86%	95%	86%	96%	90%	92%	88%	81%	68%	79%	90%	82%
85.5%	83%	88%	84%	83%	84%	84%	88%	88%	88%	91%	91%	87%	92%	91%	95%	85%	71%	65%	85%	86%	87%
81.1%	82%	80%	79%	80%	83%	87%	87%	85%	77%	91%	91%	75%	84%	86%	75%	82%	81%	73%	85%	69%	72%
83.0%	88%	94%	89%	92%	89%	94%	77%	78%	79%	88%	90%	74%	81%	82%	92%	84%	76%	65%	79%	76%	76%
85.3%	85%	92%	87%	92%	88%	81%	91%	83%	92%	91%	90%	68%	88%	70%	82%	91%	78%	93%	90%	80%	80%
75.4%	82%	80%	81%	79%	83%	72%	75%	74%	73%	81%	80%	74%	78%	82%	70%	79%	63%	65%	79%	80%	54%
76.9%	83%	82%	76%	80%	78%	61%	70%	82%	76%	77%	59%	78%	77%	70%	89%	86%	79%	81%	79%	76%	76%
81.7%	87%	92%	78%	83%	95%	71%	84%	91%	87%	82%	81%	72%	78%	70%	91%	94%	68%	83%	67%	80%	81%
72.3%	80%	87%	89%	84%	71%	66%	77%	68%	68%	68%	68%	74%	68%	64%	72%	71%	67%	68%	68%	71%	70%
77.0%	79%	89%	86%	84%	70%	80%	63%	85%	73%	76%	76%	69%	78%	65%	83%	80%	69%	65%	86%	89%	72%
84.9%	89%	86%	80%	82%	81%	88%	83%	79%	81%	88%	90%	93%	89%	93%	88%	90%	70%	76%	83%	82%	92%
72.8%	73%	79%	73%	0%	86%	70%	82%	75%	82%	70%	70%	67%	76%	73%	77%	79%	70%	78%	84%	86%	79%
59.8%	62%	60%	66%	0%	54%	69%	61%	64%	57%	68%	68%	72%	63%	66%	63%	60%	68%	57%	55%	61%	62%
80.8%	92%	79%	88%	87%	91%	72%	75%	84%	83%	85%	79%	74%	79%	74%	84%	84%	78%	82%	73%	75%	79%
73.0%	79%	86%	74%	82%	51%	79%	63%	67%	64%	62%	69%	65%	73%	79%	71%	74%	75%	73%	94%	84%	70%
82.3%	91%	88%	89%	89%	87%	81%	88%	90%	73%	88%	82%	73%	86%	73%	78%	82%	81%	70%	75%	80%	85%
89.0%	93%	93%	89%	89%	93%	91%	86%	83%	87%	92%	93%	90%	94%	94%	90%	87%	86%	79%	86%	83%	91%
89.6%	91%	91%	97%	97%	85%	97%	91%	92%	90%	97%	95%	75%	97%	87%	94%	91%	83%	79%	94%	74%	84%
83.0%	97%	86%	90%	90%	83%	73%	88%	89%	87%	94%	83%	74%	84%	76%	88%	82%	68%	74%	73%	78%	86%
89.6%	99%	93%	99%	0%	86%	98%	90%	97%	90%	98%	98%	95%	96%	99%	91%	93%	98%	90%	88%	90%	93%
82.2%	85%	92%	87%	89%	95%	80%	91%	94%	87%	89%	92%	74%	86%	82%	93%	92%	82%	37%	88%	17%	95%
72.9%	76%	76%	66%	71%	69%	64%	81%	71%	74%	78%	72%	79%	75%	76%	74%	76%	74%	72%	72%	67%	67%
72.2%	65%	84%	73%	69%	54%	80%	76%	52%	69%	70%	88%	77%	77%	79%	74%	68%	72%	82%	75%	61%	71%

Model Similarity Dendrogram

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