[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]  dndNGMI [@dndNGMI](/creator/twitter/dndNGMI) on x 1587 followers Created: 2025-06-18 16:16:59 UTC We tested three approaches to answering "what is dark matter? Models/swarm used: X. Auto-routed swarm (Qwen 2.5, DeepSeek R1 and Llama 3.3) X. DeepSeek R1 X. Lexi is uncensored The routing system successfully identified that this science question needed a quality response and selected an excellent model. But it also made X total model calls using X different models, this slowed down response time. **Summary** Auto-routed model Strengths: - Best pedagogical approach - explains what dark matter is NOT - Most accessible language while maintaining accuracy - Perfect length - comprehensive but not overwhelming - Clear structure: Evidence → Properties → Theories → Importance - Action-oriented - mentions ongoing experiments Weaknesses: - Extremely slow due to preprocessing [model:DEEPSEEK_R1] Strengths: - Shows thinking process (unique feature) - Well-structured with clear sections - Good use of formatting (bullet points, bold text) - Engaging ending with emoji - Mentions specific experiments (LUX-ZEPLIN, XENONnT) Weaknesses: - The "thinking" takes up space without adding value for the reader - Not significantly better than cheaper options - Still quite technical despite claiming to avoid jargon [model:ORENGUTENG_LLAMA_3_1_8B_LEXI_UNCENSORED_V2] Strengths: - Excellent historical context (Fritz Zwicky, 1930s) - Very comprehensive coverage - Good scientific accuracy - Well-organized with numbered points Weaknesses: - A bit dry and academic - Could be more engaging - Lacks the pedagogical clarity of the auto-routed response Like having a team of AI experts, not just one assistant  XXXXX engagements  [Post Link](https://x.com/dndNGMI/status/1935371170469064843)
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
dndNGMI @dndNGMI on x 1587 followers
Created: 2025-06-18 16:16:59 UTC
We tested three approaches to answering "what is dark matter?
Models/swarm used: X. Auto-routed swarm (Qwen 2.5, DeepSeek R1 and Llama 3.3) X. DeepSeek R1 X. Lexi is uncensored
The routing system successfully identified that this science question needed a quality response and selected an excellent model. But it also made X total model calls using X different models, this slowed down response time.
Summary
Auto-routed model Strengths:
- Best pedagogical approach - explains what dark matter is NOT
- Most accessible language while maintaining accuracy
- Perfect length - comprehensive but not overwhelming
- Clear structure: Evidence → Properties → Theories → Importance
- Action-oriented - mentions ongoing experiments
Weaknesses:
- Extremely slow due to preprocessing
[model:DEEPSEEK_R1] Strengths:
- Shows thinking process (unique feature)
- Well-structured with clear sections
- Good use of formatting (bullet points, bold text)
- Engaging ending with emoji
- Mentions specific experiments (LUX-ZEPLIN, XENONnT)
Weaknesses:
- The "thinking" takes up space without adding value for the reader
- Not significantly better than cheaper options
- Still quite technical despite claiming to avoid jargon
[model:ORENGUTENG_LLAMA_3_1_8B_LEXI_UNCENSORED_V2] Strengths:
- Excellent historical context (Fritz Zwicky, 1930s)
- Very comprehensive coverage
- Good scientific accuracy
- Well-organized with numbered points
Weaknesses:
- A bit dry and academic
- Could be more engaging
- Lacks the pedagogical clarity of the auto-routed response
Like having a team of AI experts, not just one assistant
XXXXX engagements
