Darwin Family: MRI-Trust-Weighted Evolutionary Merging for Training-Free Scaling of Language-Model Reasoning
Paper β’ 2605.14386 β’ Published β’ 60
sydgam
sydgam
AI & ML interests
None yet
Recent Activity
liked a model about 2 months ago
ginigen-ai/Rogue-27B-KR liked a model about 2 months ago
FINAL-Bench/Darwin-35B-A3B-Opus-Q8-GGUFOrganizations
None yet
upvoted a paper 17 days ago
liked 2
models about 2 months ago
liked 4
Spaces 3 months ago
MARL β Model-Agnostic Runtime Middleware for LLMs
π
24
Any LLM + 5-Stage Reasoning Pipeline
Heatmap Leaderboard korea
π₯
18
Organization Activity Heatmap
Heatmap Leaderboard
π₯
25
Organization Activity Heatmap
SiteAgent - AI μΉ μ΄μμ€ν΄νΈ
π€
22
μ΄λ€ μΉ νμ΄μ§μμλ λμνλ AI μ΄μμ€ν΄νΈ.
reacted to SeaWolf-AI's post with π€π€ππβ€οΈπ₯ 3 months ago
Post
11140
ποΈ Smol AI WorldCup: A 4B Model Just Beat 8B β Here's the Data
We evaluated 18 small language models from 12 makers on 125 questions across 7 languages. The results challenge the assumption that bigger is always better.
Community Article: https://huggingface.co/blog/FINAL-Bench/smol-worldcup
Live Leaderboard: ginigen-ai/smol-worldcup
Dataset: ginigen-ai/smol-worldcup
What we found:
β Gemma-3n-E4B (4B, 2GB RAM) outscores Qwen3-8B (8B, 5.5GB). Doubling parameters gained only 0.4 points. RAM cost: 2.75x more.
β GPT-OSS-20B fits in 1.5GB yet matches Champions-league dense models requiring 8.5GB. MoE architecture is the edge AI game-changer.
β Thinking models hurt structured output. DeepSeek-R1-7B scores 8.7 points below same-size Qwen3-8B and runs 2.7x slower.
β A 1.3B model fabricates confident fake content 80% of the time when prompted with nonexistent entities. Qwen3 family hits 100% trap detection across all sizes.
β Qwen3-1.7B (1.2GB) outscores Mistral-7B, Llama-3.1-8B, and DeepSeek-R1-14B. Latest architecture at 1.7B beats older architecture at 14B.
What makes this benchmark different?
Most benchmarks ask "how smart?" β we measure five axes simultaneously: Size, Honesty, Intelligence, Fast, Thrift (SHIFT). Our ranking metric WCS = sqrt(SHIFT x PIR_norm) rewards models that are both high-quality AND efficient. Smart but massive? Low rank. Tiny but poor? Also low.
Top 5 by WCS:
1. GPT-OSS-20B β WCS 82.6 β 1.5GB β Raspberry Pi tier
2. Gemma-3n-E4B β WCS 81.8 β 2.0GB β Smartphone tier
3. Llama-4-Scout β WCS 79.3 β 240 tok/s β Fastest model
4. Qwen3-4B β WCS 76.6 β 2.8GB β Smartphone tier
5. Qwen3-1.7B β WCS 76.1 β 1.2GB β IoT tier
Built in collaboration with the FINAL Bench research team. Interoperable with ALL Bench Leaderboard for full small-to-large model comparison.
Dataset is open under Apache 2.0 (125 questions, 7 languages). We welcome new model submissions.
We evaluated 18 small language models from 12 makers on 125 questions across 7 languages. The results challenge the assumption that bigger is always better.
Community Article: https://huggingface.co/blog/FINAL-Bench/smol-worldcup
Live Leaderboard: ginigen-ai/smol-worldcup
Dataset: ginigen-ai/smol-worldcup
What we found:
β Gemma-3n-E4B (4B, 2GB RAM) outscores Qwen3-8B (8B, 5.5GB). Doubling parameters gained only 0.4 points. RAM cost: 2.75x more.
β GPT-OSS-20B fits in 1.5GB yet matches Champions-league dense models requiring 8.5GB. MoE architecture is the edge AI game-changer.
β Thinking models hurt structured output. DeepSeek-R1-7B scores 8.7 points below same-size Qwen3-8B and runs 2.7x slower.
β A 1.3B model fabricates confident fake content 80% of the time when prompted with nonexistent entities. Qwen3 family hits 100% trap detection across all sizes.
β Qwen3-1.7B (1.2GB) outscores Mistral-7B, Llama-3.1-8B, and DeepSeek-R1-14B. Latest architecture at 1.7B beats older architecture at 14B.
What makes this benchmark different?
Most benchmarks ask "how smart?" β we measure five axes simultaneously: Size, Honesty, Intelligence, Fast, Thrift (SHIFT). Our ranking metric WCS = sqrt(SHIFT x PIR_norm) rewards models that are both high-quality AND efficient. Smart but massive? Low rank. Tiny but poor? Also low.
Top 5 by WCS:
1. GPT-OSS-20B β WCS 82.6 β 1.5GB β Raspberry Pi tier
2. Gemma-3n-E4B β WCS 81.8 β 2.0GB β Smartphone tier
3. Llama-4-Scout β WCS 79.3 β 240 tok/s β Fastest model
4. Qwen3-4B β WCS 76.6 β 2.8GB β Smartphone tier
5. Qwen3-1.7B β WCS 76.1 β 1.2GB β IoT tier
Built in collaboration with the FINAL Bench research team. Interoperable with ALL Bench Leaderboard for full small-to-large model comparison.
Dataset is open under Apache 2.0 (125 questions, 7 languages). We welcome new model submissions.
- 1 reply
upvoted an article 3 months ago
Article
ποΈ Smol AI WorldCup: A 5-Axis Benchmark That Reveals What Small Language Models Can Really Do
FINAL-Bench
β’ β’ 38 liked a
dataset 3 months ago