Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
|
@@ -16,6 +16,8 @@ ADI is a line of compact language models built at [theLAB](https://thelabsource.
|
|
| 16 |
|
| 17 |
Every model here is built end-to-end on theLAB hardware — no cloud training — then quantized to GGUF and shipped ready to run in [Ollama](https://ollama.com) or any llama.cpp-based runtime.
|
| 18 |
|
|
|
|
|
|
|
| 19 |
---
|
| 20 |
|
| 21 |
## Models
|
|
@@ -30,7 +32,7 @@ ollama run hf.co/AdvancedDataIntelligence/adi-qwen3.5-4b-glm5.2-general-GGUF:Q4_
|
|
| 30 |
|
| 31 |
### 🐱 adi-qwen3-8b-glm5.2-general
|
| 32 |
|
| 33 |
-
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,
|
| 34 |
|
| 35 |
```bash
|
| 36 |
ollama run hf.co/AdvancedDataIntelligence/adi-qwen3-8b-glm5.2-general-GGUF:Q4_K_M
|
|
@@ -70,6 +72,15 @@ Then point any llama.cpp-based runtime at the downloaded file.
|
|
| 70 |
|
| 71 |
---
|
| 72 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
## The approach
|
| 74 |
|
| 75 |
- **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).
|
|
|
|
| 16 |
|
| 17 |
Every model here is built end-to-end on theLAB hardware — no cloud training — then quantized to GGUF and shipped ready to run in [Ollama](https://ollama.com) or any llama.cpp-based runtime.
|
| 18 |
|
| 19 |
+
**Links:** [Website](https://advanced-data-intelligence.com) · [theLAB](https://thelabsource.com) · [YouTube — Advanced Data Intelligence](https://www.youtube.com/@AdvancedDataIntelligence) · [YouTube — ADI Online](https://www.youtube.com/@adi_onlin3)
|
| 20 |
+
|
| 21 |
---
|
| 22 |
|
| 23 |
## Models
|
|
|
|
| 32 |
|
| 33 |
### 🐱 adi-qwen3-8b-glm5.2-general
|
| 34 |
|
| 35 |
+
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.
|
| 36 |
|
| 37 |
```bash
|
| 38 |
ollama run hf.co/AdvancedDataIntelligence/adi-qwen3-8b-glm5.2-general-GGUF:Q4_K_M
|
|
|
|
| 72 |
|
| 73 |
---
|
| 74 |
|
| 75 |
+
## Watch
|
| 76 |
+
|
| 77 |
+
We publish short videos on AI, alignment, and how these models are built:
|
| 78 |
+
|
| 79 |
+
- **[Advanced Data Intelligence](https://www.youtube.com/@AdvancedDataIntelligence)** — the main channel.
|
| 80 |
+
- **[ADI Online](https://www.youtube.com/@adi_onlin3)** — an AI philosophy series covering the alignment problem, the paperclip maximizer, the treacherous turn, the Chinese room, and more.
|
| 81 |
+
|
| 82 |
+
---
|
| 83 |
+
|
| 84 |
## The approach
|
| 85 |
|
| 86 |
- **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).
|