minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[diffusion] chore: rename lingbot world v2 (#30518)

Browse Source
This commit is contained in:
HuangJi
2026-07-08 19:06:44 +08:00
committed by GitHub
parent 042228a195
commit 3d96bb9721
2 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs_new/cookbook/diffusion/LingBot-World/LingBot-World-2.0.mdx
+2 -2
View File
@@ -13,7 +13,7 @@ tag: REALTIME
## 1. Model Introduction ## 1. Model Introduction
LingBot World 2.0 is a realtime camera-controlled video world model. In SGLang-diffusion, it belongs to the realtime causal path: the server keeps a live session, samples control signals per chunk, reuses causal DiT state, and decodes video frames incrementally. lingbot-world-v2-14b-causal-fast-diffusers is a realtime camera-controlled video world model. In SGLang-diffusion, it belongs to the realtime causal path: the server keeps a live session, samples control signals per chunk, reuses causal DiT state, and decodes video frames incrementally.
This is different from offline diffusion video models such as Wan or LTX. Offline models denoise a bounded latent sequence for one request. Realtime world models generate a continuing stream, so the runtime must manage session state, control events, causal attention cache, and VAE decode cache. This is different from offline diffusion video models such as Wan or LTX. Offline models denoise a bounded latent sequence for one request. Realtime world models generate a continuing stream, so the runtime must manage session state, control events, causal attention cache, and VAE decode cache.
@@ -23,7 +23,7 @@ This is different from offline diffusion video models such as Wan or LTX. Offlin
export SGLANG_LINGBOT_LAZY_VAE_ENCODE_BLACK_FRAMES=60 export SGLANG_LINGBOT_LAZY_VAE_ENCODE_BLACK_FRAMES=60
export SGLANG_LINGBOT_ENABLE_INTERACTIVE_KV_WINDOW=true export SGLANG_LINGBOT_ENABLE_INTERACTIVE_KV_WINDOW=true
sglang serve \ sglang serve \
--model-path robbyant/lingbot-world-v2-diffusers \ --model-path robbyant/lingbot-world-v2-14b-causal-fast-diffusers \
--pipeline-class-name LingBotWorldCausalDMDPipeline \ --pipeline-class-name LingBotWorldCausalDMDPipeline \
--num-gpus 8 \ --num-gpus 8 \
--ulysses-degree 8 \ --ulysses-degree 8 \
python/sglang/multimodal_gen/registry.py
+1 -1
View File
@@ -769,7 +769,7 @@ def _register_configs():
sampling_param_cls=LingBotWorldSamplingParams, sampling_param_cls=LingBotWorldSamplingParams,
pipeline_config_cls=LingBotWorldV2CausalDMDConfig, pipeline_config_cls=LingBotWorldV2CausalDMDConfig,
hf_model_paths=[ hf_model_paths=[
"robbyant/lingbot-world-v2-diffusers", "robbyant/lingbot-world-v2-14b-causal-fast-diffusers",
], ],
) )
register_configs( register_configs(