From bdb0119de3bbe8a83378038a1a1e19251faa8895 Mon Sep 17 00:00:00 2001 From: Chirag Modi <98582575+cmodi-meta@users.noreply.github.com> Date: Wed, 25 Sep 2024 14:22:29 -0700 Subject: [PATCH 1/2] Update xnnpack_README.md --- .../demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md b/examples/demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md index c8b45d6c58b..aa15eb8f34c 100644 --- a/examples/demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md +++ b/examples/demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md @@ -49,7 +49,7 @@ sh examples/models/llama2/install_requirements.sh ### For Llama 3.2 1B and 3B models We have supported BFloat16 as a data type on the XNNPack backend for Llama 3.2 1B/3B models. -* You can download original model weights for Llama through Meta official [website](https://llama.meta.com/), or via Huggingface (Link to specific 3.2 1B repo) +* You can download original model weights for Llama through Meta official [website](https://llama.meta.com/). * For chat use-cases, download the instruct models instead of pretrained. * Run “examples/models/llama2/install_requirements.sh” to install dependencies. * The 1B model in BFloat16 format can run on mobile devices with 8GB RAM (iPhone 15 Pro and later). The 3B model will require 12GB+ RAM and hence will not fit on 8GB RAM phones. From ac8c2ee7baa4c92b88ed12ae7020203697867171 Mon Sep 17 00:00:00 2001 From: cmodi-meta <98582575+cmodi-meta@users.noreply.github.com> Date: Wed, 25 Sep 2024 14:40:29 -0700 Subject: [PATCH 2/2] minor change for 3.2 link in readme --- .../demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md b/examples/demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md index aa15eb8f34c..73b1e614f63 100644 --- a/examples/demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md +++ b/examples/demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md @@ -1,6 +1,6 @@ # Building Llama iOS Demo for XNNPack Backend -**[UPDATE - 09/25]** We have added support for running [Llama 3.2 models](#for-llama-3.2-1b-and-3b-models) on the XNNPack backend. We currently support inference on their original data type (BFloat16). +**[UPDATE - 09/25]** We have added support for running [Llama 3.2 models](#for-llama-32-1b-and-3b-models) on the XNNPack backend. We currently support inference on their original data type (BFloat16). This tutorial covers the end to end workflow for building an iOS demo app using XNNPack backend on device. More specifically, it covers: