From 6d5bf9564cb444758cbe58dc4319d872b3fdd496 Mon Sep 17 00:00:00 2001 From: Teeeio <> Date: Thu, 22 Jan 2026 21:35:37 +0800 Subject: [PATCH 1/2] Fix outdated installation instructions for Emu3.5 Update installation method to use vLLM-FL backend instead of the removed patch mechanism, aligning with the v1.0.0-alpha.0 refactor that moved hardware-specific support to plugin repositories. Changes: - Remove obsolete tools/patch/unpatch.py command - Update to use vLLM-FL backend installation - Align with docs/getting-started.md instructions Fixes: Documentation inconsistency after v1.0.0-alpha.0 refactor --- examples/emu3.5/conf/README.md | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/examples/emu3.5/conf/README.md b/examples/emu3.5/conf/README.md index 2c541d2149..3f2721f33b 100644 --- a/examples/emu3.5/conf/README.md +++ b/examples/emu3.5/conf/README.md @@ -5,14 +5,16 @@ vLLM implementation of https://github.com/baaivision/Emu3.5 ## Environment Setup ### Install FlagScale -- Build from source code base on vLLM tag/0.11.0 +- Install FlagScale and vLLM-FL backend (based on vLLM) ```bash git clone https://github.com/flagos-ai/FlagScale.git -cd FlgScale -python tools/patch/unpatch.py --backend vllm -cd FlagScale/third_party/vllm -pip install -r requirements/cuda.txt --no-cache-dir -MAX_JOBS=32 pip install --no-build-isolation -v . +cd FlagScale +pip install . --verbose + +# Install vLLM-FL backend +git clone https://github.com/flagos-ai/vllm-FL +cd vllm-FL +pip install -e . ``` ### Prepare Emu3.5 From 6826c94b0811b451b8416e1812e922dcd7fd5946 Mon Sep 17 00:00:00 2001 From: Teeeio <> Date: Thu, 22 Jan 2026 16:31:46 +0000 Subject: [PATCH 2/2] docs: fix repository URLs and config file paths in model documentation - Update repository URLs from FlagOpen/FlagScale to flagos-ai/FlagScale across all model docs - Fix incorrect config file path in robobrain2_5 (3b.yaml -> 32b.yaml) Aligns with repository migration to flagos-ai organization. --- examples/pi0/README.md | 2 +- examples/pi0_5/README.md | 2 +- examples/qwen2_5_vl/README.md | 2 +- examples/qwen3_vl/README.md | 4 ++-- examples/robobrain2/README.md | 2 +- examples/robobrain2_5/README.md | 4 ++-- examples/robobrain_x0/README.md | 2 +- examples/robobrain_x0_5/README.md | 2 +- 8 files changed, 10 insertions(+), 10 deletions(-) diff --git a/examples/pi0/README.md b/examples/pi0/README.md index 13f02422dd..fd19332790 100644 --- a/examples/pi0/README.md +++ b/examples/pi0/README.md @@ -7,7 +7,7 @@ This guide covers how to train, run inference, and serve PI0 models using FlagSc ### Clone Repository ```sh -git clone https://github.com/FlagOpen/FlagScale.git +git clone https://github.com/flagos-ai/FlagScale.git cd FlagScale/ ``` diff --git a/examples/pi0_5/README.md b/examples/pi0_5/README.md index e345d25e45..a4dede25b7 100644 --- a/examples/pi0_5/README.md +++ b/examples/pi0_5/README.md @@ -7,7 +7,7 @@ This guide covers how to train, run inference, and serve PI0.5 models using Flag ### Clone Repository ```sh -git clone https://github.com/FlagOpen/FlagScale.git +git clone https://github.com/flagos-ai/FlagScale.git cd FlagScale/ ``` diff --git a/examples/qwen2_5_vl/README.md b/examples/qwen2_5_vl/README.md index c5954444f5..e85b55f648 100644 --- a/examples/qwen2_5_vl/README.md +++ b/examples/qwen2_5_vl/README.md @@ -5,7 +5,7 @@ ### Clone Repository ```sh -git clone https://github.com/FlagOpen/FlagScale.git +git clone https://github.com/flagos-ai/FlagScale.git cd FlagScale/ ``` diff --git a/examples/qwen3_vl/README.md b/examples/qwen3_vl/README.md index 47bc4c26df..f2ad54b191 100644 --- a/examples/qwen3_vl/README.md +++ b/examples/qwen3_vl/README.md @@ -7,7 +7,7 @@ Download the source code。 ```bash -git clone https://github.com/FlagOpen/FlagScale.git +git clone https://github.com/flagos-ai/FlagScale.git cd FlagScale ``` @@ -24,7 +24,7 @@ pip install --no-build-isolation . pip install transformers==4.57.1 ``` -You can also refer to the readme in `https://github.com/FlagOpen/FlagScale.git` +You can also refer to the readme in `https://github.com/flagos-ai/FlagScale.git` ### 2. Prepare checkpoint diff --git a/examples/robobrain2/README.md b/examples/robobrain2/README.md index e15305aa33..5f72b4fdc6 100644 --- a/examples/robobrain2/README.md +++ b/examples/robobrain2/README.md @@ -6,7 +6,7 @@ ### Clone Repository ```sh -git clone https://github.com/FlagOpen/FlagScale.git +git clone https://github.com/flagos-ai/FlagScale.git cd FlagScale/ ``` diff --git a/examples/robobrain2_5/README.md b/examples/robobrain2_5/README.md index e9f87b9a34..0494de2f79 100644 --- a/examples/robobrain2_5/README.md +++ b/examples/robobrain2_5/README.md @@ -6,7 +6,7 @@ ### Clone Repository ```sh -git clone https://github.com/FlagOpen/FlagScale.git +git clone https://github.com/flagos-ai/FlagScale.git cd FlagScale/ ``` @@ -87,7 +87,7 @@ tail -f outputs/robobrain2.5_4b/serve_logs/host_0_localhost.output ```sh cd FlagScale/ -vim examples/robobrain2_5/conf/serve/3b.yaml +vim examples/robobrain2_5/conf/serve/32b.yaml ``` Change 1 fields: diff --git a/examples/robobrain_x0/README.md b/examples/robobrain_x0/README.md index f5766f5973..9c486ca2a3 100644 --- a/examples/robobrain_x0/README.md +++ b/examples/robobrain_x0/README.md @@ -3,7 +3,7 @@ ### Clone Repository ```sh -git clone https://github.com/FlagOpen/FlagScale.git +git clone https://github.com/flagos-ai/FlagScale.git cd FlagScale/ ``` diff --git a/examples/robobrain_x0_5/README.md b/examples/robobrain_x0_5/README.md index 18d8ea2cac..522dd0f05f 100644 --- a/examples/robobrain_x0_5/README.md +++ b/examples/robobrain_x0_5/README.md @@ -3,7 +3,7 @@ ### Clone Repository ```sh -git clone https://github.com/FlagOpen/FlagScale.git +git clone https://github.com/flagos-ai/FlagScale.git cd FlagScale/ ```