From 022165a3c824afcec9b67169ff7cfc26543d4c34 Mon Sep 17 00:00:00 2001 From: yangrui6 Date: Fri, 24 Apr 2026 14:49:13 +0800 Subject: [PATCH] docs(deepeyes): remove missing run_deepeyes_4b.sh refs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit # 📝 Documentation ## Drop references to non-existent 4B launch script - Remove "4B Model (8 GPUs)" quick-start section from en/zh docs - Swap model download snippet to `Qwen3-VL-30B-A3B-Thinking` - Update Ray Job example to invoke `run_deepeyes.sh` - Drop `run_deepeyes_4b.sh` entry from file-structure listing --- docs/en/examples/deepeyes.md | 20 +++----------------- docs/zh/examples/deepeyes.md | 20 +++----------------- 2 files changed, 6 insertions(+), 34 deletions(-) diff --git a/docs/en/examples/deepeyes.md b/docs/en/examples/deepeyes.md index fe19547..ce065c0 100644 --- a/docs/en/examples/deepeyes.md +++ b/docs/en/examples/deepeyes.md @@ -49,25 +49,12 @@ The HF Image dict format (`{"bytes": ...}`) is natively supported by Relax's ima ### Download the Model ```bash -hf download Qwen/Qwen3-VL-4B-Instruct \ - --local-dir /root/Qwen3-VL-4B-Instruct +hf download Qwen/Qwen3-VL-30B-A3B-Thinking \ + --local-dir /root/Qwen3-VL-30B-A3B-Thinking ``` -For the full-scale configuration, use `Qwen/Qwen3-VL-30B-A3B-Thinking`. - ## Quick Start -### 4B Model (8 GPUs) - -```bash -export MODEL_DIR=/root -export DATA_DIR=/root -export SAVE_DIR=/root/save - -cd /root/Relax -bash examples/deepeyes/run_deepeyes_4b.sh -``` - ### 30B-A3B Model (8 GPUs, MoE) The full-scale configuration requires a judge model for reward scoring: @@ -91,7 +78,7 @@ bash examples/deepeyes/run_deepeyes.sh ```bash WORKING_DIR="./" RAY_ADDRESS=:6379 \ MODEL_DIR=/root DATA_DIR=/root SAVE_DIR=/root/save \ - bash -x scripts/entrypoint/ray-job.sh examples/deepeyes/run_deepeyes_4b.sh + bash -x scripts/entrypoint/ray-job.sh examples/deepeyes/run_deepeyes.sh ``` ## Architecture @@ -101,7 +88,6 @@ WORKING_DIR="./" RAY_ADDRESS=:6379 \ ``` examples/deepeyes/ ├── run_deepeyes.sh # Launch script (Qwen3-VL-30B-A3B, full config) -├── run_deepeyes_4b.sh # Launch script (Qwen3-VL-4B, lightweight) ├── deepeyes_config.yaml # Task config (max_turns, env path) ├── rollout.py # Multi-turn rollout logic ├── env_deepeyes.py # DeepEyes tool-use environment diff --git a/docs/zh/examples/deepeyes.md b/docs/zh/examples/deepeyes.md index 950b151..959799f 100644 --- a/docs/zh/examples/deepeyes.md +++ b/docs/zh/examples/deepeyes.md @@ -49,25 +49,12 @@ HF Image dict 格式（`{"bytes": ...}`）被 Relax 的图像加载管线原生 ### 下载模型 ```bash -hf download Qwen/Qwen3-VL-4B-Instruct \ - --local-dir /root/Qwen3-VL-4B-Instruct +hf download Qwen/Qwen3-VL-30B-A3B-Thinking \ + --local-dir /root/Qwen3-VL-30B-A3B-Thinking ``` -完整配置使用 `Qwen/Qwen3-VL-30B-A3B-Thinking`。 - ## 快速开始 -### 4B 模型（8 GPU） - -```bash -export MODEL_DIR=/root -export DATA_DIR=/root -export SAVE_DIR=/root/save - -cd /root/Relax -bash examples/deepeyes/run_deepeyes_4b.sh -``` - ### 30B-A3B 模型（8 GPU，MoE） 完整配置需要 judge 模型进行奖励评分： @@ -91,7 +78,7 @@ bash examples/deepeyes/run_deepeyes.sh ```bash WORKING_DIR="./" RAY_ADDRESS=:6379 \ MODEL_DIR=/root DATA_DIR=/root SAVE_DIR=/root/save \ - bash -x scripts/entrypoint/ray-job.sh examples/deepeyes/run_deepeyes_4b.sh + bash -x scripts/entrypoint/ray-job.sh examples/deepeyes/run_deepeyes.sh ``` ## 架构 @@ -101,7 +88,6 @@ WORKING_DIR="./" RAY_ADDRESS=:6379 \ ``` examples/deepeyes/ ├── run_deepeyes.sh # 启动脚本（Qwen3-VL-30B-A3B，完整配置） -├── run_deepeyes_4b.sh # 启动脚本（Qwen3-VL-4B，轻量配置） ├── deepeyes_config.yaml # 任务配置（max_turns、环境路径） ├── rollout.py # 多轮 rollout 逻辑 ├── env_deepeyes.py # DeepEyes 工具使用环境