Cross-link: DistilQwen collection spotlight — 2026-03-29
Browse files
README.md
CHANGED
|
@@ -203,3 +203,19 @@ Not intended for:
|
|
| 203 |
|
| 204 |
|
| 205 |
*Last updated: 2026-03-28 12:58 UTC*
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 203 |
|
| 204 |
|
| 205 |
*Last updated: 2026-03-28 12:58 UTC*
|
| 206 |
+
|
| 207 |
+
<!-- CIX-CROSSLINK-START -->
|
| 208 |
+
|
| 209 |
+
---
|
| 210 |
+
|
| 211 |
+
## From the Convergent Intelligence Portfolio
|
| 212 |
+
|
| 213 |
+
**[DistilQwen Collection](https://huggingface.co/collections/reaperdoesntknow/distilqwen-69bf40ec669117e3f069ef1c)** — Proof-weighted distillation from Qwen3-30B-A3B → 1.7B and 0.6B. Three teacher variants (Instruct, Thinking, Coder), nine models, 2,788 combined downloads. Structure beats scale.
|
| 214 |
+
|
| 215 |
+
Top model: [Qwen3-1.7B-Coder-Distilled-SFT](https://huggingface.co/reaperdoesntknow/Qwen3-1.7B-Coder-Distilled-SFT) — 508 downloads
|
| 216 |
+
|
| 217 |
+
Full methodology: [Structure Over Scale (DOI: 10.57967/hf/8165)](https://doi.org/10.57967/hf/8165)
|
| 218 |
+
|
| 219 |
+
*Convergent Intelligence LLC: Research Division*
|
| 220 |
+
|
| 221 |
+
<!-- CIX-CROSSLINK-END -->
|