deploy: a56bea0

Author: zhuzilin

.doctrees/environment.pickle

@@ -54,7 +54,7 @@
     <link rel="prev" title="Search-R1 lite" href="../search-r1/README.html" />
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-    <meta name="docbuild:last-update" content="Sep 08, 2025"/>
+    <meta name="docbuild:last-update" content="Sep 09, 2025"/>
   </head>
 
 
@@ -592,7 +592,7 @@ <h2>Config Differences (2 Key Points)<a class="headerlink" href="#config-differe
 
   <div class="footer-item">
     <p class="last-updated">
-  Last updated on Sep 08, 2025.
+  Last updated on Sep 09, 2025.
   <br/>
 </p>
   </div>

.doctrees/get_started/quick_start.doctree

@@ -54,7 +54,7 @@
     <link rel="prev" title="Retool: from SFT to RL" href="../retool/README.html" />
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-    <meta name="docbuild:last-update" content="Sep 08, 2025"/>
+    <meta name="docbuild:last-update" content="Sep 09, 2025"/>
   </head>
 
 
@@ -580,7 +580,7 @@ <h2>New Arguments<a class="headerlink" href="#new-arguments" title="Link to this
 
   <div class="footer-item">
     <p class="last-updated">
-  Last updated on Sep 08, 2025.
+  Last updated on Sep 09, 2025.
   <br/>
 </p>
   </div>

_examples_synced/fully_async/README.html

@@ -54,7 +54,7 @@
     <link rel="prev" title="Fully Asynchronous Rollout Example" href="../fully_async/README.html" />
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-    <meta name="docbuild:last-update" content="Sep 08, 2025"/>
+    <meta name="docbuild:last-update" content="Sep 09, 2025"/>
   </head>
 
 
@@ -639,7 +639,7 @@ <h2>Safety Features<a class="headerlink" href="#safety-features" title="Link to
 
   <div class="footer-item">
     <p class="last-updated">
-  Last updated on Sep 08, 2025.
+  Last updated on Sep 09, 2025.
   <br/>
 </p>
   </div>

_examples_synced/multi_agent/README.html

@@ -54,7 +54,7 @@
     <link rel="prev" title="SFT Qwen3-4B-Base" href="../../examples/qwen3-4b-base-openhermes.html" />
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-    <meta name="docbuild:last-update" content="Sep 08, 2025"/>
+    <meta name="docbuild:last-update" content="Sep 09, 2025"/>
   </head>
 
 
@@ -594,7 +594,7 @@ <h2>Code Structure<a class="headerlink" href="#code-structure" title="Link to th
 
   <div class="footer-item">
     <p class="last-updated">
-  Last updated on Sep 08, 2025.
+  Last updated on Sep 09, 2025.
   <br/>
 </p>
   </div>

_examples_synced/retool/README.html

@@ -130,7 +130,8 @@ CKPT_ARGS=(
    --hf-checkpoint /root/GLM-Z1-9B-0414
    # Reference Model's Megatron format checkpoint
    --ref-load /root/GLM-Z1-9B-0414_torch_dist
-   # Actor model loading path. If empty, load from --ref-load
+   # Actor model loading path. Should typically match --save for checkpoint resumption
+   # If empty or doesn't contain a valid checkpoint, loads from --ref-load instead
    --load /root/GLM-Z1-9B-0414_slime/
    # Model save path during training
    --save /root/GLM-Z1-9B-0414_slime/
@@ -185,7 +186,7 @@ ROLLOUT_ARGS=(
    --n-samples-per-prompt 8
    --num-steps-per-rollout 1
    --global-batch-size 128
-   
+
    # Rollout sampling parameters
    --rollout-max-response-len 8192
    --rollout-temperature 0.8
@@ -490,7 +491,7 @@ async def generate(args, sample: Sample, sampling_params) -> Sample:
             # ... tokenization and appending ...
             loss_masks += [0] * len(tool_tokens) # loss_mask = 0
             full_response += tool_output
-            
+
         elif action == "answer":
             break # end loop
 
@@ -553,4 +554,4 @@ ray job submit --address="http://127.0.0.1:8265" \
 slime has been deeply optimized for distributed training of large-scale Mixture of Experts (MoE) models. We provide some end-to-end training cases for reference:
 
 - [Example: 64xH100 Training GLM-4.5](models/glm4.5-355B-A32B.md)
-- [Example: 128xH100 Training DeepSeek-R1](models/deepseek-r1.md) 
+- [Example: 128xH100 Training DeepSeek-R1](models/deepseek-r1.md)

_examples_synced/search-r1/README.html

@@ -54,7 +54,7 @@
     <link rel="prev" title="Multi-Agent RL" href="../_examples_synced/multi_agent/README.html" />
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-    <meta name="docbuild:last-update" content="Sep 08, 2025"/>
+    <meta name="docbuild:last-update" content="Sep 09, 2025"/>
   </head>
 
 
@@ -629,7 +629,7 @@ <h3><a class="reference external" href="https://github.com/sgl-project/sglang/pu
 
   <div class="footer-item">
     <p class="last-updated">
-  Last updated on Sep 08, 2025.
+  Last updated on Sep 09, 2025.
   <br/>
 </p>
   </div>

_sources/get_started/quick_start.md

@@ -53,7 +53,7 @@
     <link rel="prev" title="v0.1.0: Redefining High-Performance RL Training Frameworks" href="release_v0.1.0.html" />
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-    <meta name="docbuild:last-update" content="Sep 08, 2025"/>
+    <meta name="docbuild:last-update" content="Sep 09, 2025"/>
   </head>
 
 
@@ -625,7 +625,7 @@ <h2>Roadmap<a class="headerlink" href="#roadmap" title="Link to this heading">#<
 
   <div class="footer-item">
     <p class="last-updated">
-  Last updated on Sep 08, 2025.
+  Last updated on Sep 09, 2025.
   <br/>
 </p>
   </div>

advanced/speculative-decoding.html

@@ -54,7 +54,7 @@
     <link rel="prev" title="AMD" href="../platform_support/amd_tutorial.html" />
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-    <meta name="docbuild:last-update" content="Sep 08, 2025"/>
+    <meta name="docbuild:last-update" content="Sep 09, 2025"/>
   </head>
 
 
@@ -683,7 +683,7 @@ <h2>Correctness Verification<a class="headerlink" href="#correctness-verificatio
 
   <div class="footer-item">
     <p class="last-updated">
-  Last updated on Sep 08, 2025.
+  Last updated on Sep 09, 2025.
   <br/>
 </p>
   </div>