Agora board builds (Model Farm, verified on real boards)

NOTICE.md

@@ -0,0 +1,5 @@
+# Provenance
+
+Resources are copied without binary changes from the APLUX Model Farm build `QuickSRNetMedium` that Agora verified on real boards (decoded sample check, then measured), pinned by sha256 in SHA256SUMS.
+
+Agora arranges files under targets/<SoC>/<precision>/ and adds model.json. AidLite, AidGen, AidVoice and the Qualcomm QNN runtime are separate dependencies with their own terms.

README.md

@@ -0,0 +1,38 @@
+---
+license: other
+license_name: aplux-model-farm
+license_link: https://huggingface.co/APLUX-Agora/QuickSRNetMedium/blob/main/LICENSE_MODEL_FARM.pdf
+pipeline_tag: image-to-image
+tags: [Agora, QNN, AIoT, Qualcomm]
+---
+
+# QuickSRNet 4× model resources
+
+高通 QuickSRNet，把低分辨率图片放大 4 倍并恢复细节。
+
+Board builds that [Agora](https://mega.aidlux.com) runs and measures on real Qualcomm boards. Try it online, without a board: https://mega.aidlux.com/models/quicksrnet-medium
+
+| SoC · board | Precisions | Measured on Agora (default precision) |
+| --- | --- | --- |
+| QCS6490 · 犀牛派 A1 (Rhino Pi A1) | W8A16, W8A8 | 2.1 ms |
+| QCS8550 · 犀牛派 X1 (Rhino Pi X1) | FP16, W8A16, W8A8 | 1.7 ms |
+
+Task: 图像增强. Inputs: image (image). Outputs: image (image).
+
+## Download
+
+```bash
+export AGORA_WORKSPACE="$HOME/my-quicksrnet-medium"
+export AGORA_SOC='QCS8550'
+hf download APLUX-Agora/QuickSRNetMedium \
+  --include "targets/$AGORA_SOC/*" --local-dir "$AGORA_WORKSPACE/resources"
+```
+
+Each `targets/<SoC>/` holds `model.json` (default precision and variant paths) and one directory per precision. Context binaries are named `model.bin`; multi-file packages are kept as published by Model Farm.
+
+Pre- and post-processing, the runner and an exported project: open the model on Agora and choose “部署指南”, or export any workflow that uses it.
+
+## Licence and provenance
+
+Source model licence: see the original model's terms. 
+Converted, deployable resources: APLUX Model Farm License (LICENSE_MODEL_FARM.pdf). See NOTICE.md and SHA256SUMS.

SHA256SUMS

@@ -0,0 +1,5 @@
+a7086667ff7621e30a57c34601cabe7e202f98a2cde7078add2c092d5a3fdd47  targets/QCS6490/W8A16/model.bin
+15327f2ecc7d350a2e0f03666909eb868f0365d326022753cf6ab9ae0fd77189  targets/QCS6490/W8A8/model.bin
+be2b71517d279701d6c6d81892bcc2c87d8397ab188622533bbdbbd10acde514  targets/QCS8550/FP16/model.bin
+0434a9c1839123bb836f4bda74aad6f846a1c2dae3c7c43b71477e926d9ef5f2  targets/QCS8550/W8A16/model.bin
+098f132381c707293924a6a24ca40bf43d0bdfbafb583daec333b7d2bcbbc49f  targets/QCS8550/W8A8/model.bin

targets/QCS6490/W8A16/model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7086667ff7621e30a57c34601cabe7e202f98a2cde7078add2c092d5a3fdd47
+size 119496

targets/QCS6490/W8A8/model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15327f2ecc7d350a2e0f03666909eb868f0365d326022753cf6ab9ae0fd77189
+size 119496

targets/QCS6490/model.json

@@ -0,0 +1,8 @@
+{
+  "soc": "QCS6490",
+  "default_precision": "W8A16",
+  "variants": {
+    "W8A16": "W8A16/model.bin",
+    "W8A8": "W8A8/model.bin"
+  }
+}

targets/QCS8550/FP16/model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be2b71517d279701d6c6d81892bcc2c87d8397ab188622533bbdbbd10acde514
+size 185320

targets/QCS8550/W8A16/model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0434a9c1839123bb836f4bda74aad6f846a1c2dae3c7c43b71477e926d9ef5f2
+size 119496

targets/QCS8550/W8A8/model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:098f132381c707293924a6a24ca40bf43d0bdfbafb583daec333b7d2bcbbc49f
+size 119496

targets/QCS8550/model.json

@@ -0,0 +1,9 @@
+{
+  "soc": "QCS8550",
+  "default_precision": "FP16",
+  "variants": {
+    "FP16": "FP16/model.bin",
+    "W8A16": "W8A16/model.bin",
+    "W8A8": "W8A8/model.bin"
+  }
+}