v0.60.0
Browse filesSee https://github.com/qualcomm/ai-hub-models/releases/v0.60.0 for changelog.
- README.md +3 -3
- release_assets.json +3 -3
README.md
CHANGED
|
@@ -16,7 +16,7 @@ pipeline_tag: text-generation
|
|
| 16 |
Phi-3.5-mini is a lightweight, state-of-the-art open model built upon datasets used for Phi-3 - synthetic data and filtered publicly available websites - with a focus on very high-quality, reasoning dense data. The model belongs to the Phi-3 model family and supports 128K token context length. The model underwent a rigorous enhancement process, incorporating both supervised fine-tuning, proximal policy optimization, and direct preference optimization to ensure precise instruction adherence and robust safety measures.
|
| 17 |
|
| 18 |
This is based on the implementation of Phi-3.5-Mini-Instruct found [here](https://huggingface.co/microsoft/Phi-3.5-mini-instruct).
|
| 19 |
-
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.
|
| 20 |
|
| 21 |
Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
|
| 22 |
|
|
@@ -29,8 +29,8 @@ Below are pre-exported model assets ready for deployment.
|
|
| 29 |
|
| 30 |
| Runtime | Precision | Chipset | SDK Versions | Download |
|
| 31 |
|---|---|---|---|---|
|
| 32 |
-
| GENIE | w4a16 | Snapdragon® X2 Elite | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/phi_3_5_mini_instruct/releases/v0.
|
| 33 |
-
| GENIE | w4a16 | Snapdragon® X Elite | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/phi_3_5_mini_instruct/releases/v0.
|
| 34 |
|
| 35 |
For more device-specific assets and performance metrics, visit **[Phi-3.5-Mini-Instruct on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/phi_3_5_mini_instruct)**.
|
| 36 |
|
|
|
|
| 16 |
Phi-3.5-mini is a lightweight, state-of-the-art open model built upon datasets used for Phi-3 - synthetic data and filtered publicly available websites - with a focus on very high-quality, reasoning dense data. The model belongs to the Phi-3 model family and supports 128K token context length. The model underwent a rigorous enhancement process, incorporating both supervised fine-tuning, proximal policy optimization, and direct preference optimization to ensure precise instruction adherence and robust safety measures.
|
| 17 |
|
| 18 |
This is based on the implementation of Phi-3.5-Mini-Instruct found [here](https://huggingface.co/microsoft/Phi-3.5-mini-instruct).
|
| 19 |
+
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/phi_3_5_mini_instruct) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).
|
| 20 |
|
| 21 |
Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device.
|
| 22 |
|
|
|
|
| 29 |
|
| 30 |
| Runtime | Precision | Chipset | SDK Versions | Download |
|
| 31 |
|---|---|---|---|---|
|
| 32 |
+
| GENIE | w4a16 | Snapdragon® X2 Elite | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/phi_3_5_mini_instruct/releases/v0.60.0/phi_3_5_mini_instruct-genie-w4a16-qualcomm_snapdragon_x2_elite.zip)
|
| 33 |
+
| GENIE | w4a16 | Snapdragon® X Elite | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/phi_3_5_mini_instruct/releases/v0.60.0/phi_3_5_mini_instruct-genie-w4a16-qualcomm_snapdragon_x_elite.zip)
|
| 34 |
|
| 35 |
For more device-specific assets and performance metrics, visit **[Phi-3.5-Mini-Instruct on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/phi_3_5_mini_instruct)**.
|
| 36 |
|
release_assets.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"version": "0.
|
| 3 |
"precisions": {
|
| 4 |
"w4a16": {
|
| 5 |
"chipset_assets": {
|
|
@@ -8,7 +8,7 @@
|
|
| 8 |
"tool_versions": {
|
| 9 |
"qairt": "2.43.1.260218"
|
| 10 |
},
|
| 11 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/phi_3_5_mini_instruct/releases/v0.
|
| 12 |
}
|
| 13 |
},
|
| 14 |
"qualcomm-snapdragon-x2-elite": {
|
|
@@ -16,7 +16,7 @@
|
|
| 16 |
"tool_versions": {
|
| 17 |
"qairt": "2.43.1.260218"
|
| 18 |
},
|
| 19 |
-
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/phi_3_5_mini_instruct/releases/v0.
|
| 20 |
}
|
| 21 |
}
|
| 22 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"version": "0.60.0",
|
| 3 |
"precisions": {
|
| 4 |
"w4a16": {
|
| 5 |
"chipset_assets": {
|
|
|
|
| 8 |
"tool_versions": {
|
| 9 |
"qairt": "2.43.1.260218"
|
| 10 |
},
|
| 11 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/phi_3_5_mini_instruct/releases/v0.60.0/phi_3_5_mini_instruct-genie-w4a16-qualcomm_snapdragon_x_elite.zip"
|
| 12 |
}
|
| 13 |
},
|
| 14 |
"qualcomm-snapdragon-x2-elite": {
|
|
|
|
| 16 |
"tool_versions": {
|
| 17 |
"qairt": "2.43.1.260218"
|
| 18 |
},
|
| 19 |
+
"download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/phi_3_5_mini_instruct/releases/v0.60.0/phi_3_5_mini_instruct-genie-w4a16-qualcomm_snapdragon_x2_elite.zip"
|
| 20 |
}
|
| 21 |
}
|
| 22 |
}
|