Spaces:
Running
Running
| title: README | |
| emoji: 😻 | |
| colorFrom: yellow | |
| colorTo: blue | |
| sdk: static | |
| pinned: false | |
| short_description: Small, local models distilled from frontier teachers | |
| <p align="center"> | |
| <img src="http://serve.thelabsource.com/u/FhQgYP.gif" width="720" alt="Advanced Data Intelligence"> | |
| </p> | |
| <h1 align="center">Advanced Data Intelligence</h1> | |
| <p align="center"> | |
| <strong>Small, local, open models — distilled from frontier teachers.</strong> | |
| </p> | |
| <p align="center"> | |
| ADI is a line of compact language models built at <a href="https://thelabsource.com">theLAB</a> (<em>Learning. Algorithms. Breakthroughs.</em>). Each model is a <strong>knowledge distillation</strong>: a strong frontier "teacher" generates high-quality answers across thousands of prompts, and a small "student" model is fine-tuned to imitate them — producing a model that reasons and responds like something much larger, while staying small enough to run on a single consumer GPU. | |
| </p> | |
| <p align="center"> | |
| Every model here is built end-to-end on theLAB hardware — no cloud training — then quantized to GGUF and shipped ready to run in <a href="https://ollama.com">Ollama</a> or any llama.cpp-based runtime. | |
| </p> | |
| <p align="center"> | |
| <strong>Links:</strong> | |
| <a href="https://advanced-data-intelligence.com">Website</a> · | |
| <a href="https://thelabsource.com">theLAB</a> · | |
| <a href="https://www.youtube.com/@AdvancedDataIntelligence">YouTube — Advanced Data Intelligence</a> · | |
| <a href="https://www.youtube.com/@adi_onlin3">YouTube — ADI Online</a> | |
| </p> | |
| <p align="center"> | |
| <img src="https://serve.thelabsource.com/u/PjCf8w.png" width="700" alt=""> | |
| </p> | |
| <p align="center"> | |
| <img src="https://serve.thelabsource.com/u/oLHfvT.png" width="900" alt="New here? Start with one of these — adi-qwen3.5-4b, adi-qwen2.5-coder-7b, adi-qwen3-8b"> | |
| </p> | |
| ### 🐱 adi-qwen3.5-4b-glm5.2-general | |
| General-purpose local assistant. Qwen3.5-4B distilled from **glm-5.2**. Reasons and explains like a frontier model on general topics. Native tool-calling, 262K context, ~2.7 GB. | |
| ```bash | |
| ollama run hf.co/AdvancedDataIntelligence/adi-qwen3.5-4b-glm5.2-general-GGUF:Q4_K_M | |
| ``` | |
| ### 🐱 adi-qwen3-8b-glm5.2-general | |
| General-purpose local assistant. Qwen3-8B distilled from **glm-5.2**. Reasons and explains like a frontier model on general topics, with more headroom than the 4B. Native tool-calling, 128K context, ~5 GB. | |
| ```bash | |
| ollama run hf.co/AdvancedDataIntelligence/adi-qwen3-8b-glm5.2-general-GGUF:Q4_K_M | |
| ``` | |
| ### 🐱 adi-qwen3.5-9b-glm5.2-general | |
| General-purpose local assistant. Qwen3.5-9B distilled from **glm-5.2**. The most capable general student in the line — more parametric headroom for nuanced reasoning while still fitting a single consumer GPU. Native tool-calling, 262K context, ~5.6 GB. | |
| ```bash | |
| ollama run hf.co/AdvancedDataIntelligence/adi-qwen3.5-9b-glm5.2-general-GGUF:Q4_K_M | |
| ``` | |
| ### 🐱 adi-qwen2.5-coder-7b-kimi2.7-code | |
| Local coding assistant. Qwen2.5-Coder-7B distilled from **kimi-k2.7-code**. Writes, explains, and debugs code with frontier-style quality. Native tool-calling, 128K context, ~4.4 GB. | |
| ```bash | |
| ollama run hf.co/AdvancedDataIntelligence/adi-qwen2.5-coder-7b-kimi2.7-code-GGUF:Q4_K_M | |
| ``` | |
| <p align="center"> | |
| <img src="https://serve.thelabsource.com/u/uSv2Lp.png" width="760" alt="ADI model lineup — size on disk"> | |
| </p> | |
| --- | |
| <h2 align="center">Browse the whole line</h2> | |
| <p align="center"> | |
| <strong><a href="https://huggingface.co/spaces/AdvancedDataIntelligence/adi-models-lab">ADI Models Lab</a></strong> — the full lineup in one place. Pick a student from the rail (Qwen3.5 4B, Qwen3.5 9B, Qwen3 8B, Coder 7B, and the <code>hey-adi</code> wakeword), read its teacher, context, and size at a glance, then copy a ready-to-paste run command. Includes the live in-browser demo — no install to try, no sign-in to copy. | |
| </p> | |
| <p align="center"> | |
| <a href="https://huggingface.co/spaces/AdvancedDataIntelligence/adi-models-lab"> | |
| <img src="http://serve.thelabsource.com/u/c8cTr3.gif" alt="ADI Models Lab — pick a student, copy a command, run offline" width="800"> | |
| </a> | |
| </p> | |
| <p align="center"> | |
| <em>Pick a student. Copy a command. Run offline.</em><br> | |
| <a href="https://huggingface.co/spaces/AdvancedDataIntelligence/adi-models-lab">▶ Open ADI Models Lab</a> | |
| </p> | |
| --- | |
| <h2 align="center">Try it live</h2> | |
| <p align="center"> | |
| A hosted demo is available as a Hugging Face Space — chat with the model directly in your browser, no install required. | |
| </p> | |
| <p align="center"> | |
| <a href="https://huggingface.co/spaces/AdvancedDataIntelligence/adi-qwen3.5-4b-glm5.2-general-demo"> | |
| <img src="https://serve.thelabsource.com/u/4Kb3iS.gif" alt="adi-qwen3.5-4b-glm5.2-general live demo" width="800"> | |
| </a> | |
| </p> | |
| <p align="center"> | |
| <a href="https://huggingface.co/spaces/AdvancedDataIntelligence/adi-qwen3.5-4b-glm5.2-general-demo">▶ Launch the demo</a> | |
| </p> | |
| --- | |
| ## How to run | |
| **Ollama (recommended).** Pull and run any model directly from this org — no manual download needed. Ollama fetches the GGUF from Hugging Face on first run: | |
| ```bash | |
| ollama run hf.co/AdvancedDataIntelligence/adi-qwen3-8b-glm5.2-general-GGUF:Q4_K_M | |
| ``` | |
| Swap `:Q4_K_M` for another quant tag if a model ships multiple. To pull without running: | |
| ```bash | |
| ollama pull hf.co/AdvancedDataIntelligence/adi-qwen3-8b-glm5.2-general-GGUF:Q4_K_M | |
| ``` | |
| **Manual download (llama.cpp or offline).** Grab the raw GGUF with the Hugging Face CLI: | |
| ```bash | |
| huggingface-cli download AdvancedDataIntelligence/adi-qwen3-8b-glm5.2-general-GGUF adi-qwen3-8b-glm5.2-q4_k_m.gguf --local-dir . | |
| ``` | |
| Then point any llama.cpp-based runtime at the downloaded file. | |
| <p align="center"> | |
| <img src="https://serve.thelabsource.com/u/T5Kdlg.png" width="700" alt=""> | |
| </p> | |
| ## The approach | |
| - **Distillation, not retraining.** We transfer a teacher's reasoning style and answer quality into a small student — not net-new facts. For raw recall, pair these with retrieval (RAG). | |
| - **Local-first.** Every model runs fully offline on consumer hardware. No API, no data leaving the machine. | |
| - **Open.** Apache-2.0 where the base license allows, with full training details on each model card so the work is reproducible. | |
| <p align="center"> | |
| <img src="https://serve.thelabsource.com/u/ih5dUC.png" width="760" alt="The ADI distillation pipeline"> | |
| </p> | |
| --- | |
| ## Coming next | |
| In the pipeline, distilled the same way and headed here soon: | |
| - **adi-qwen2.5-14b-glm5.2-general** — a larger general student with more parametric headroom. | |
| - **adi-gemma3-12b-glm5.2-general** — a Gemma-based general distill, broadening the lineup beyond Qwen. | |
| Follow the org to catch them on release. | |
| ## Naming | |
| Models follow the pattern `adi-<base>-<size>-<teacher>-<purpose>` — so the name tells you the student base, its size, the teacher it learned from, and what it's tuned for. | |
| <p align="center"> | |
| <img src="https://serve.thelabsource.com/u/yZylMt.gif" width="560" alt="ADI"> | |
| </p> | |
| <p align="center"> | |
| <img src="https://serve.thelabsource.com/u/O8Cq1i.gif" width="560" alt="ADI"> | |
| </p> | |
| <p align="center"> | |
| <em>Built at <a href="https://thelabsource.com">theLAB</a> — Learning. Algorithms. Breakthroughs.</em> | |
| </p> |