Instructions to use thoddnn/chat-adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use thoddnn/chat-adapter with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download thoddnn/chat-adapter --local-dir chat-adapter
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Publish verified incremental chat checkpoint 126 from production step 1900
Browse files- README.md +5 -1
- adapters.safetensors +1 -1
- manifest.json +6 -4
README.md
CHANGED
|
@@ -11,6 +11,10 @@ tags:
|
|
| 11 |
|
| 12 |
Qwen 3.5 2B coordinator adapter for Fenn’s shared chat SDK. This adapter generates bounded Python tool calls; it is not a general document-analysis adapter.
|
| 13 |
|
| 14 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
Source base: [mlx-community/Qwen3.5-2B-8bit](https://huggingface.co/mlx-community/Qwen3.5-2B-8bit). The base retains its upstream Apache 2.0 license and attribution.
|
|
|
|
| 11 |
|
| 12 |
Qwen 3.5 2B coordinator adapter for Fenn’s shared chat SDK. This adapter generates bounded Python tool calls; it is not a general document-analysis adapter.
|
| 13 |
|
| 14 |
+
Current release: incremental SFT checkpoint 126, continued from production checkpoint 1900. Rank 16; four adapted layers; 8-bit base. This release adds indexing and duplicate-scope demonstrations with replay examples.
|
| 15 |
+
|
| 16 |
+
Use the matching Nano base revision in `manifest.json` and Fenn’s current tool harness. Load the base without this adapter when using Nano as an analyst. Manifest checksums cover the adapter configuration and weights.
|
| 17 |
+
|
| 18 |
+
The adapter passed integrity and fresh-process reload checks. A five-case focused tool-use evaluation improved from one to four passing cases; this small evaluation is not a general quality benchmark. Duplicate scanning without an explicit folder can still produce an internal workspace path, which the current Fenn harness redirects to known indexed folders.
|
| 19 |
|
| 20 |
Source base: [mlx-community/Qwen3.5-2B-8bit](https://huggingface.co/mlx-community/Qwen3.5-2B-8bit). The base retains its upstream Apache 2.0 license and attribution.
|
adapters.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 11220564
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1ddba57b493965c5e05ef7d6808fb31e6d48fe6d385a931f3f5cad8cacceda1
|
| 3 |
size 11220564
|
manifest.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"schema": "fenn-chat-adapter-v1",
|
| 3 |
-
"id": "
|
| 4 |
"base_repo": "thoddnn/Qwen3.5-2B-8bit",
|
| 5 |
"base_revision": "dc098b420dc52743e621579391d40ef26e2b9d06",
|
| 6 |
"source_repo": "mlx-community/Qwen3.5-2B-8bit",
|
| 7 |
"source_revision": "562fb8ea19bbe7565a3ddf5b7d49899ea2c88d2b",
|
| 8 |
"quantization_bits": 8,
|
| 9 |
-
"checkpoint":
|
| 10 |
"files": {
|
| 11 |
"adapter_config.json": {
|
| 12 |
"bytes": 132,
|
|
@@ -14,7 +14,9 @@
|
|
| 14 |
},
|
| 15 |
"adapters.safetensors": {
|
| 16 |
"bytes": 11220564,
|
| 17 |
-
"sha256": "
|
| 18 |
}
|
| 19 |
-
}
|
|
|
|
|
|
|
| 20 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"schema": "fenn-chat-adapter-v1",
|
| 3 |
+
"id": "production-scope-incremental-2026-11-14",
|
| 4 |
"base_repo": "thoddnn/Qwen3.5-2B-8bit",
|
| 5 |
"base_revision": "dc098b420dc52743e621579391d40ef26e2b9d06",
|
| 6 |
"source_repo": "mlx-community/Qwen3.5-2B-8bit",
|
| 7 |
"source_revision": "562fb8ea19bbe7565a3ddf5b7d49899ea2c88d2b",
|
| 8 |
"quantization_bits": 8,
|
| 9 |
+
"checkpoint": 126,
|
| 10 |
"files": {
|
| 11 |
"adapter_config.json": {
|
| 12 |
"bytes": 132,
|
|
|
|
| 14 |
},
|
| 15 |
"adapters.safetensors": {
|
| 16 |
"bytes": 11220564,
|
| 17 |
+
"sha256": "b1ddba57b493965c5e05ef7d6808fb31e6d48fe6d385a931f3f5cad8cacceda1"
|
| 18 |
}
|
| 19 |
+
},
|
| 20 |
+
"parent_checkpoint": 1900,
|
| 21 |
+
"parent_adapter_sha256": "7394bb3f6a9f8674b98559f4f6d244fb6ea65bc5ccf2c748fcf239f954989df8"
|
| 22 |
}
|