TroyDoesAI/Codestral-RAG-19B-Pruned
Text Generation • 19B • Updated • 11 • 2
Troy Schultz PRO
TroyDoesAI
AI & ML interests
Contact Me ~ Open For Work ~ Contract or W2
Recent Activity
reacted to SeaWolf-AI's post with 🔥 21 days ago
🧬 Darwin-35B-A3B-Opus — The Child That Surpassed Both Parents
What if a merged model could beat both its parents? We proved it can.
Darwin-35B-A3B-Opus is a 35B MoE model (3B active) built with our Darwin V5 engine — the first evolution system that CT-scans parent models before merging them.
🤗 Model: https://huggingface.co/FINAL-Bench/Darwin-35B-A3B-Opus
The result speaks for itself: GPQA Diamond 90.0%, versus Father (Qwen3.5-35B-A3B) at 84.2% and Mother (Claude 4.6 Opus Distilled) at 85.0%. That's +6.9% over Father and +5.9% over Mother. Not a tradeoff — a genuine leap. Meanwhile, MMMLU sits at 85.0% (Father: 85.2%), multimodal is fully intact, and all 201 languages are preserved.
How? Model MRI changed everything. Traditional merging is guesswork. Darwin V4 added evolution. Darwin V5 added X-ray vision. Model MRI scans each parent layer by layer and discovers: Mother's L34–L38 is the reasoning engine (peak cosine distance), 50–65% of Mother's experts are dead (killed by text-only distillation), and Father is a healthy generalist with every expert alive. The prescription: transplant Mother's reasoning brain at L38 (90% weight), replace her dead experts with Father's living ones, and let Father's router handle the output layer. Reasoning went up. Versatility stayed intact. No tradeoff — just evolution.
35B total, 3B active (MoE) · GPQA Diamond 90.0% · MMMLU 85.0% (201 languages) · Multimodal Image & Video · 262K native context · 147.8 tok/s on H100 · Runs on a single RTX 4090 (Q4) · Apache 2.0
Darwin V5's full algorithm and technical details will be released alongside an upcoming paper.
🚀 Live Demo: https://huggingface.co/spaces/FINAL-Bench/Darwin-35B-A3B-Opus
🏆 FINAL Bench Leaderboard: https://huggingface.co/spaces/FINAL-Bench/Leaderboard
📊 ALL Bench Leaderboard: https://huggingface.co/spaces/FINAL-Bench/all-bench-leaderboard
Built by VIDRAFT · Supported by the Korean Government GPU Support Program new activity 21 days ago
blascotobasco/Mistral-NeMoE-12B-16E:Wow what a cool experiment liked a model 21 days ago
blascotobasco/Mistral-NeMoE-12B-16EOrganizations
TroyDoesAI/Codestral-21B-Pruned
Text Generation • 21B • Updated • 15 • 2
TroyDoesAI/CodeStral-RAG-Lora
Updated • 1
TroyDoesAI/Codestral-22B-RAG-Q8-gguf
22B • Updated • 45 • 11
TroyDoesAI/Contextual-Obedient-MoE-3x8B-Llama3-RAG
Text Generation • 19B • Updated • 12 • 1
TroyDoesAI/Contextual-Llama3-8B-RAG
Text Generation • 8B • Updated • 3 • 1
TroyDoesAI/Mermaid-Contextual-Obedient-RAG-Phi-3-medium-128k-instruct-18B
Text Generation • 19B • Updated • 4 • 2
TroyDoesAI/Contexual-Llama-18B-RAG
Text Generation • 19B • Updated • 1
TroyDoesAI/Contextual-Llama-13B-RAG
Text Generation • 13B • Updated • 1
TroyDoesAI/TinyLlama-RAG
Text Generation • 1B • Updated • 1
TroyDoesAI/Phi-3-Context-Obedient-RAG-7B
Text Generation • 7B • Updated • 8
TroyDoesAI/Tiny-RAG-gguf
1B • Updated
TroyDoesAI/Phi-3-Context-Obedient-RAG
Text Generation • 4B • Updated • 21 • 31
TroyDoesAI/Mermaid-Llama-3-3B-Pruned
Text Generation • 4B • Updated • 4
TroyDoesAI/Mermaid-Llama-6.7B-RAG-Code-Instruct
Text Generation • 7B • Updated • 16 • 1
TroyDoesAI/Mermaid-Llama-6.7B-RAG
Text Generation • 7B • Updated • 16 • 23
TroyDoesAI/Contextual-Llama-6.7b
Text Generation • 7B • Updated • 1
TroyDoesAI/Mermaid-Llama-3-5B-Pruned
Text Generation • 5B • Updated • 10 • 1
TroyDoesAI/Mermaid-Llama-3-6B-Pruned
Text Generation • 6B • Updated • 7
TroyDoesAI/Mermaid-Llama-3-7B-Pruned
Text Generation • 7B • Updated • 15 • 1
TroyDoesAI/Mermaid-Llama-3-8B
Text Generation • 8B • Updated • 13 • 11
TroyDoesAI/Mermaid-Dolphin-Mixtral-2x7b
Text Generation • 13B • Updated • 10 • 1
TroyDoesAI/MermaidSolar_LASER
Text Generation • 11B • Updated • 11 • 1
TroyDoesAI/Mermaid-Can-Code
Text Generation • 8B • Updated • 3
TroyDoesAI/Mermaid-Llama-3-8B-SNR-LASER
Text Generation • 8B • Updated • 5
TroyDoesAI/Mermaid-Llama-3-4B-Pruned
Text Generation • 5B • Updated • 5 • 1
TroyDoesAI/MermaidMoE-19B
Text Generation • 19B • Updated • 13 • 1
TroyDoesAI/Mermaid-Flow-MoE-Expert2
Text Generation • 11B • Updated • 10
TroyDoesAI/Mermaid-Coder-MoE-Expert1
Text Generation • 11B • Updated • 5
TroyDoesAI/Mermaid-Solar
Text Generation • 11B • Updated • 8