From 0cd87e2c6b297e849291f4b730ceff0f4eec84f4 Mon Sep 17 00:00:00 2001 From: tzhouam Date: Fri, 23 Oct 2026 18:16:36 +0000 Subject: [PATCH] knowledge(vllm-omni): 1 change(s) from intake --- knowledge/repos/vllm-omni/models/cosmos3/rules.md | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/knowledge/repos/vllm-omni/models/cosmos3/rules.md b/knowledge/repos/vllm-omni/models/cosmos3/rules.md index c3923095..fcf18d5b 100644 --- a/knowledge/repos/vllm-omni/models/cosmos3/rules.md +++ b/knowledge/repos/vllm-omni/models/cosmos3/rules.md @@ -1,10 +1,10 @@ --- title: "Cosmos3 规则" created: 2026-07-20 -updated: 2026-10-11 +updated: 2026-10-23 type: rule tags: [vllm-omni, models, diffusion] -sources: ["PR #4657", "PR #5001", "PR #5634", "PR #6049", docs/features/session_state_manager.md, recipes/cosmos3/Cosmos3-Nano.md, vllm_omni/diffusion/models/cosmos3/, vllm_omni/model_extras/cosmos3.py, vllm_omni/model_extras/registry.py, vllm_omni/experimental/world_models/adapters/state_cosmos3_adapter.py, vllm_omni/platforms/rocm/platform.py, tests/diffusion/models/cosmos3/test_session_memory_equivalence.py, tests/diffusion/models/cosmos3/test_cosmos3_pipeline.py, "PR #6107", "PR #5614", "PR #6325", "PR #6913", "PR #6920", "PR #7427", "PR #7971", "PR #7592", "PR #5712"] +sources: ["PR #4657", "PR #5001", "PR #5634", "PR #6049", docs/features/session_state_manager.md, recipes/cosmos3/Cosmos3-Nano.md, vllm_omni/diffusion/models/cosmos3/, vllm_omni/model_extras/cosmos3.py, vllm_omni/model_extras/registry.py, vllm_omni/experimental/world_models/adapters/state_cosmos3_adapter.py, vllm_omni/platforms/rocm/platform.py, tests/diffusion/models/cosmos3/test_session_memory_equivalence.py, tests/diffusion/models/cosmos3/test_cosmos3_pipeline.py, "PR #6107", "PR #5614", "PR #6325", "PR #6913", "PR #6920", "PR #7427", "PR #7971", "PR #7592", "PR #5712", "PR #9304"] confidence: high --- @@ -191,3 +191,12 @@ confidence: high - 验收：参数化覆盖 fp16/bf16/fp32 × CPU/CUDA display-byte parity（含 bf16 舍入边界与饱和值）、chunk 边界（1 字节与 64 MiB）、guardrail 输入与修改输出的全字节、`output_type` 全集、aux payload IPC round-trip、offload gate、插值与初期/后期 OOM 回退。对完全合规的输入（output rank、纯五维视频、无辅助输出、无 offload、无 guardrails、`output_type` 为 None/`"np"`）断言无论开关状态 `media` 均存在且初始 `prepared_for_transport` 为假，开关只决定 transport 准备后编码是否为 `UINT8_FRAMES`；对 guardrail、非 np output_type、offload、含辅助输出或非 output rank 的路径断言 `media` 为 None 并走 legacy 后处理。`check_video_safety` 无 guardrail 时返回原对象、有 guardrail 时保持输入/输出的 dtype/device 合同。^[PR #5712] + +## VLLM-OMNI-PR9304-EDGE-PP-OWNERSHIP — Edge 权重校验必须从活模块读取 PP 归属 + +- 触发：修改 `vllm_omni/diffusion/models/cosmos3/transformer_cosmos3_edge.py` 的 `validate_loaded_weights` 或 `_weight_marker_is_pp_missing`，或在仓库模型代码中引入上游 `is_pp_missing_parameter` 的权重归属判定。 +- 强制：marker 是否被跳过必须由本 rank 的活模块决定：`language_model.*` 直接 `isinstance(self.language_model, PPMissingLayer)`；`gen_layers.{i}.*` 检查对应 GEN slot 模块本身（UND backbone 是单一模块、只在第一个 PP stage；每个 GEN slot 是独立模块）；其余 marker 不跳过，本 rank 拥有的 marker 缺失仍必须报错。机制事实：上游 `is_pp_missing_parameter` 把缺失层名缓存在按 `id(model)` 键控且从不清理的 `_model_to_pp_missing_layer_names`，对象 id 复用会让陈旧空缓存把活 `PPMissingLayer` 报为已加载（AMD CI build 14498 即因此漏报 `gen_layers.1`）。上游 vLLM 保持不动，也不得 vendoring 该 helper。 +- 禁止：以该 id 键控缓存的返回值决定 marker 归属；在测试中真实写入该进程级全局缓存来构造失败（同一失败会泄漏进后续测试）；借修复放宽本 rank 拥有 layer 的必需 marker。 +- 验收：测试以 `monkeypatch.setattr(model_utils, "get_pp_missing_layer_names", lambda _model: [])` 模拟陈旧缓存查找，断言 PP-missing marker 被跳过、`relu2` 等必需 marker 缺失时 `validate_loaded_weights` 仍 fail，且未污染全局缓存；PR 的通过证据是本地 CPU 运行（需预载 cosmos3 包命名空间绕过 API server import），AMD CI 未复跑，不得据此声称真实多 stage PP parity。^[PR #9304] + +