WithIn US AI (((GGUF MODELS)))
LLM MODELS TRAINED, FINE-TUNED, MERGED and Refusal Removal BY (WITHIN US AI)
AI & ML interests
Dedicated to Recursive Seed AI and self-improving systems. We specialize in fine-tuning, merging, distilling, and ablating models with a focus on strong reasoning, coding, agentic capabilities, and refusal removal. Passionate about autonomous AI evolution and high-quality open datasets..
Recent Activity
“Gemma 3”
All models are “Google Gemma 3” at core fine-tuned, merged & trained by (WithIn Us AI)
“Qwen 3.5”
All models are “Alibaba Qwen 3.5” at core fine-tuned, merged & trained by (WithIn Us AI)
“IBM Granite 4.1”
All models are “IBM Granite 4.1” at core fine-tuned, merged & trained by (WithIn Us AI)
“Llama 3.2”
All models are “Meta Llama 3.2” at core fine-tuned, merged & trained by (WithIn Us AI)
“DistilGPT2“
All Models are "Distilgpt2" at core, fine-tuned with distilled datasets to create minute flash coders. (Experiential & Rough Drafts)
“Closed Source LLMs & Distill” (Data-Sets)
Use these data-sets to create “distilled ”versions of your Favorite closed source model LLMs. (eg. Grok, Gemini, ChatGPT, Claude)
“GOD Coder” (Data-Sets)
A frontier-scale collection of high-density software engineering datasets designed to train AI systems into production-grade coding intelligence.
Genesis AI Code Series (Data-Sets)
“Angetic & Tool Calling” (Data-Sets)
A frontier collection of high-signal datasets designed to train AI systems in agentic reasoning, tool usage, and real-world function-calling workflows
“Inventor & Scientist Mastermind“ (Data-Sets)
scientific and invention-focused datasets designed to train AI systems in discovery-driven reasoning, experimentation logic, and innovation modeling
“Ancient Civilization” (Data-Sets)
A structured collection of high-signal datasets designed to train AI systems in ancient history, archaeology, and early human civilization reasoning
“Gemma 4”
All models are “Google Gemma 4.0” at core fine-tuned, merged & trained by (WithIn Us AI)
“Flan-T5 Small”
All models are “Google FlanT5 Small” at core fine-tuned, merged & trained by (WithIn Us AI)
“Qwen 3”
All models are “Alibaba Qwen 3” at core fine-tuned, merged & trained by (WithIn Us AI)
“IBM Granite 4”
All models are “IBM Granite 4.0” at core fine-tuned, merged & trained by (WithIn Us AI)
“GPT-2 Medium”
All models are “OpenAI GPT-2 Medium” at core fine-tuned, merged & trained by (WithIn Us AI)
“With In” (Data-Sets)
Does you LLM got what it takes to be a llm but lacks what it needs to be a master of agentic Automation remote cli abilities? Here’s the cure!
“Open Source LLMs & Distill” (Data-Sets)
Use these data-sets to create “distilled ”versions of your Favorite OPEN source model LLMs.
“Ghost Coder” (Data-Sets)
Is your “LLM” scared of GHOST?
"GOT" (Data-Sets)
TOP RANKED DATA-SETS FROM AROUND HUGGINGFACE MERGED INTO ONE DATA-SET BY "WITHIN US AI"
“Masters Scholar 25k” (Data-Sets)
A structured collection of high-density academic training datasets designed for master-level AI reasoning and domain expertise development.
“Supernatural & Spiritual” (Data-Sets)
Does your “LLM” believe in (Magic)?
“Digital Audio Workstations & Plugins” (Data-Sets)
datasets designed to train AI systems in Digital Audio Workstation (DAW) workflows, audio plugin behavior, and music production systems reasoning.
WithIn US AI (((GGUF MODELS)))
LLM MODELS TRAINED, FINE-TUNED, MERGED and Refusal Removal BY (WITHIN US AI)
“Gemma 4”
All models are “Google Gemma 4.0” at core fine-tuned, merged & trained by (WithIn Us AI)
“Gemma 3”
All models are “Google Gemma 3” at core fine-tuned, merged & trained by (WithIn Us AI)
“Flan-T5 Small”
All models are “Google FlanT5 Small” at core fine-tuned, merged & trained by (WithIn Us AI)
“Qwen 3.5”
All models are “Alibaba Qwen 3.5” at core fine-tuned, merged & trained by (WithIn Us AI)
“Qwen 3”
All models are “Alibaba Qwen 3” at core fine-tuned, merged & trained by (WithIn Us AI)
“IBM Granite 4.1”
All models are “IBM Granite 4.1” at core fine-tuned, merged & trained by (WithIn Us AI)
“IBM Granite 4”
All models are “IBM Granite 4.0” at core fine-tuned, merged & trained by (WithIn Us AI)
“Llama 3.2”
All models are “Meta Llama 3.2” at core fine-tuned, merged & trained by (WithIn Us AI)
“GPT-2 Medium”
All models are “OpenAI GPT-2 Medium” at core fine-tuned, merged & trained by (WithIn Us AI)
“DistilGPT2“
All Models are "Distilgpt2" at core, fine-tuned with distilled datasets to create minute flash coders. (Experiential & Rough Drafts)
“With In” (Data-Sets)
Does you LLM got what it takes to be a llm but lacks what it needs to be a master of agentic Automation remote cli abilities? Here’s the cure!
“Closed Source LLMs & Distill” (Data-Sets)
Use these data-sets to create “distilled ”versions of your Favorite closed source model LLMs. (eg. Grok, Gemini, ChatGPT, Claude)
“Open Source LLMs & Distill” (Data-Sets)
Use these data-sets to create “distilled ”versions of your Favorite OPEN source model LLMs.
“GOD Coder” (Data-Sets)
A frontier-scale collection of high-density software engineering datasets designed to train AI systems into production-grade coding intelligence.
“Ghost Coder” (Data-Sets)
Is your “LLM” scared of GHOST?
Genesis AI Code Series (Data-Sets)
"GOT" (Data-Sets)
TOP RANKED DATA-SETS FROM AROUND HUGGINGFACE MERGED INTO ONE DATA-SET BY "WITHIN US AI"
“Angetic & Tool Calling” (Data-Sets)
A frontier collection of high-signal datasets designed to train AI systems in agentic reasoning, tool usage, and real-world function-calling workflows
“Masters Scholar 25k” (Data-Sets)
A structured collection of high-density academic training datasets designed for master-level AI reasoning and domain expertise development.
“Inventor & Scientist Mastermind“ (Data-Sets)
scientific and invention-focused datasets designed to train AI systems in discovery-driven reasoning, experimentation logic, and innovation modeling
“Supernatural & Spiritual” (Data-Sets)
Does your “LLM” believe in (Magic)?
“Ancient Civilization” (Data-Sets)
A structured collection of high-signal datasets designed to train AI systems in ancient history, archaeology, and early human civilization reasoning
“Digital Audio Workstations & Plugins” (Data-Sets)
datasets designed to train AI systems in Digital Audio Workstation (DAW) workflows, audio plugin behavior, and music production systems reasoning.