From decf231edc2c33d84a8a107e8522415e23ee4bca Mon Sep 17 00:00:00 2001
From: limeng <lm18532679755@163.com>
Date: Fri, 24 Apr 2026 10:45:38 +0000
Subject: [PATCH] docs: add FlashSAC to support matrix and algorithms doc

---
 docs/users/zh_CN/02-simulation-backends.md |  2 ++
 docs/users/zh_CN/04-algorithms.md          | 35 ++++++++++++++++++++++
 src/unilab/docs/support_matrix.py          |  7 +++++
 3 files changed, 44 insertions(+)

diff --git a/docs/users/zh_CN/02-simulation-backends.md b/docs/users/zh_CN/02-simulation-backends.md
index e798e88b7..01d83e2ae 100644
--- a/docs/users/zh_CN/02-simulation-backends.md
+++ b/docs/users/zh_CN/02-simulation-backends.md
@@ -71,6 +71,8 @@ uv run scripts/generate_support_matrix.py --write
 | SAC (torch) | `g1_walk_rough` (G1 walk rough) | Tested | Registered |
 | SAC (torch) | `g1_sac_wbt` (g1 sac wbt) | Tested | - |
 | TD3 (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Registered |
+| FlashSAC (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Tested |
+| FlashSAC (torch) | `g1_walk_flat_amp` (g1 walk flat amp) | Tested | Registered |
 
 ### Source Index
 
diff --git a/docs/users/zh_CN/04-algorithms.md b/docs/users/zh_CN/04-algorithms.md
index fa24b2dcb..89735fa1a 100644
--- a/docs/users/zh_CN/04-algorithms.md
+++ b/docs/users/zh_CN/04-algorithms.md
@@ -116,6 +116,41 @@ uv run scripts/train_offpolicy.py algo=td3 task=td3/g1_walk_flat/mujoco training
 | `training.play_only` | false | 仅回放 |
 | `training.no_play` | false | 跳过自动回放 |
 
+## FlashSAC
+
+FlashSAC 是基于 FlashAttention 风格 Block 的 off-policy 算法，actor 使用 BatchNorm embedder + 结构化噪声探索，critic 使用 distributional Q（C51 变体）。与 FastSAC 共用同一个训练入口 `train_offpolicy.py`，但网络结构和前向接口不同。
+
+### Usage
+
+```bash
+# 基本训练
+uv run scripts/train_offpolicy.py algo=flashsac task=flashsac/g1_walk_flat/mujoco
+
+# 跳过自动回放
+uv run scripts/train_offpolicy.py algo=flashsac task=flashsac/g1_walk_flat/mujoco training.no_play=true
+```
+
+### Playback
+
+```bash
+uv run scripts/train_offpolicy.py algo=flashsac task=flashsac/g1_walk_flat/mujoco training.play_only=true
+uv run scripts/train_offpolicy.py algo=flashsac task=flashsac/g1_walk_flat/mujoco training.play_only=true algo.load_run="2026-04-23_14-06-57_mujoco"
+```
+
+### Key Parameters
+
+| 参数 | 默认值 | 说明 |
+|------|--------|------|
+| `algo` | `flashsac` | 算法选择 |
+| `task` | `flashsac/g1_walk_flat/mujoco` | 唯一 owner 配置入口；reward/env/algo 都在这里改 |
+| `algo.max_iterations` | 5000 | 最大训练迭代数 |
+| `algo.num_envs` | 1024 | 并行环境数量 |
+| `algo.tau` | 0.01 | target network 软更新系数 |
+| `algo.algo_params.actor_num_blocks` | 2 | actor FlashSAC block 层数 |
+| `algo.algo_params.critic_num_blocks` | 2 | critic FlashSAC block 层数 |
+| `training.play_only` | false | 仅回放 |
+| `training.no_play` | false | 跳过自动回放 |
+
 ## Navigation
 
 - Previous: [Training Guide](03-training.md)
diff --git a/src/unilab/docs/support_matrix.py b/src/unilab/docs/support_matrix.py
index 5fe9931f4..7a9e4a355 100644
--- a/src/unilab/docs/support_matrix.py
+++ b/src/unilab/docs/support_matrix.py
@@ -118,6 +118,13 @@ class SupportRow:
         task_glob="*/*.yaml",
         generic_tested=True,
     ),
+    EntrypointSpec(
+        entrypoint_id="flashsac_torch",
+        label="FlashSAC (torch)",
+        config_dir="conf/offpolicy/task/flashsac",
+        task_glob="*/*.yaml",
+        generic_tested=True,
+    ),
 )