[Bug Report] Severe VRAM Allocation Instability in PyTorch after llama-cpp-python is Imported
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
- 难度
- 5/5
- 预计耗时
- 一周以上
- 新手友好度
- 30/100
- Issue 类型
- 缺陷
- 描述清晰度
- 基本清楚
- 活跃度
- 停滞
调研方向
该 issue 描述了 llama-cpp-python 与 PyTorch 之间的底层 CUDA context 冲突。首先检查 llama-cpp-python 的 bindings 中的 CUDA 初始化代码,尤其是 import 时附近的代码。查找任何 CUDA context 创建操作或内存分配器设置。在一个会 import llama-cpp-python、随后加载大型 PyTorch 模型的最小环境中复现该问题,同时监控 VRAM 分配模式。“Done”意味着找出导致冲突的具体调用或配置,并提出修复方案。
由索引模型根据 Issue 内容生成。
描述
Prerequisites
- I am running the latest code.
- I carefully followed the README.md.
- I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
- I reviewed the Discussions, and have a new bug to share.
Expected Behavior
When loading a large PyTorch-based model (e.g., an SDXL checkpoint >10GB) within a host application (ComfyUI), I expect the GPU's dedicated memory (VRAM) usage to increase smoothly and stabilize at the required level for the model. This process should be fast and efficient.
Current Behavior
The VRAM allocation is extremely erratic and inefficient. The memory usage graph shows a severe sawtooth/spiky pattern—it spikes up, drops slightly, spikes again, and repeats this cycle until the model is finally loaded. This process is significantly slower than normal and suggests a fundamental conflict in memory management at the CUDA level.
This behavior only occurs when a component that imports llama-cpp-python is loaded by the host application at startup. The issue manifests before any llama-cpp-python functions are actively called; the mere act of importing the library is enough to trigger this conflict with subsequent PyTorch operations.
Environment and Context
-
Hardware:
- GPU: NVIDIA RTX PRO 6000 Blackwell Workstation Edition (94.5 GB VRAM)
-
Operating System:
- Windows 11
-
SDK and Software Versions:
- Python Version:
3.12.11 - Compiler: MSVC / Visual Studio
2022used to compile for CUDA - CUDA Toolkit: 12.8 (locally compiled)
- NVIDIA Driver Version:
580.97 - Host Application: ComfyUI (Release
2025-08-27or newer)
- Python Version:
-
Key Python Packages:
- llama-cpp-python:
0.3.16 - torch:
2.8.0+cu128 - diskcache: 5.6.3
- llama-cpp-python:
Failure Information (for bugs)
This report details a deep-seated conflict between llama-cpp-python and torch regarding CUDA context and/or VRAM management.
Investigation Summary & Key Findings
We conducted an extensive, multi-day investigation to isolate this issue, and the results were conclusive:
-
The problem is triggered by the
importofllama-cpp-python. Any host application that loads this library at startup will cause subsequent, unrelatedtorchVRAM allocations to become unstable. Disabling the components that importllama-cpp-pythoncompletely resolves the issue. -
The
diskcachedependency is a Red Herring. We initially suspected thediskcachelibrary was the cause. To test this, we performed a definitive experiment: we hid the realdiskcachelibrary and replaced it with a completely inert, non-functional fake Python package that only contained an emptyCacheclass.- Result: The VRAM instability problem persisted even with the fake library.
- Log Confirmation: The application log confirmed our fake library was being used:
--- 警告:正在使用一個偽造的 (FAKE) diskcache 函式庫 --- Set up system temp directory: C:\Users\Win\AppData\Local\Temp\latentsync_16880d62 - Conclusion: This proves that
diskcache's code is not the active cause. The problem isllama-cpp-pythonitself. The initial correlation was due todiskcachebeing a mandatory dependency; uninstalling it simply preventedllama-cpp-pythonfrom being imported successfully.
Final Hypothesis
The evidence overwhelmingly suggests a low-level conflict between how llama-cpp-python (especially a version compiled against a modern CUDA 12.8 toolkit) initializes its CUDA context and how PyTorch expects to manage the VRAM. The initialization of llama-cpp-python appears to "pollute" or "fragment" the VRAM state, making subsequent large-scale allocations by PyTorch highly inefficient.
Steps to Reproduce
- Set up a ComfyUI environment with PyTorch and the system components listed above.
- Compile and install
llama-cpp-pythonwith CUDA 12.8 support. - Install any ComfyUI custom node that imports
llama-cpp-pythonat startup (e.g.,ComfyUI-JoyCaptionorComfyUI-Searge-LLM-Node). - Launch ComfyUI. The server will start, and
llama-cpp-pythonwill be imported into the process. - Load a standard workflow designed to load a large PyTorch-based checkpoint model (e.g., an SDXL checkpoint, >10GB).
- Observe the "Dedicated GPU Memory" graph in Windows Task Manager. You will see the erratic, spiky allocation pattern instead of a smooth ramp-up.
Note on llama.cpp comparison: This issue is not about the performance or functionality of llama-cpp-python's inference, but rather a resource conflict it creates with another major library (torch). Therefore, comparing behavior against llama.cpp's ./main executable is not applicable, as the issue requires both runtimes to be active in the same Python process.
- 主要语言
- Python
- 星标
- 10.7k
- 派生
- 1.5k
- 平均合并
- 1 小时 40 分钟
- 30 天内合并 PR
- 5
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 没有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue,再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库,在一个分支上完成修改。
- 提交 Pull Request,并在描述里引用这个 Issue 编号。
abetlen/llama-cpp-python 的其他 Issue
-
难度 2/5 1-3 小时 新手友好度 78/100
abetlen/llama-cpp-python#2424 ·
维护者通常 1 天内回复
-
Seven llama_sampler_init_* bindings admit keyword arguments that the ctypes function object silently drops可能已有人在做 @Belal0066 于 59 天前认领。 未关闭
难度 2/5 1-3 小时 新手友好度 88/100
abetlen/llama-cpp-python#2371 ·
维护者通常 1 天内回复
-
uv add llama-cpp-python wheels fails for versions above 0.3.30可能已有人在做 关联的 PR 仍在进行中或已合并。 未关闭
难度 2/5 1-3 小时 新手友好度 65/100
abetlen/llama-cpp-python#2352 · 1 条评论 · 3 个 reaction ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 75/100
abetlen/llama-cpp-python#2314 ·
维护者通常 1 天内回复
-
Llama.embed() calls LlamaBatch.add_sequence with old 3-arg signature; missing logits_array可能重新可做 @lxcxjxhx 于 136 天前认领,目前没有进行中的 PR。 未关闭
难度 2/5 1-3 小时 新手友好度 65/100
abetlen/llama-cpp-python#2211 · 2 条评论 ·
维护者通常 1 天内回复
查看 abetlen/llama-cpp-python 的全部 Issue
相似的 Issue
-
Bug IronPython Python 3
难度 2/5 1-3 小时 新手友好度 72/100
pyrevitlabs/pyRevit#4184 ·
维护者通常 1 天内回复
-
难度 1/5 1 小时以内 新手友好度 74/100
jqueryscript/awesome-claude-code#1009 ·
-
tests
难度 2/5 1-3 小时 新手友好度 66/100
维护者通常 3 天内回复
-
难度 2/5 1-3 小时 新手友好度 72/100
维护者通常 1 天内回复
-
难度 1/5 1-3 小时 新手友好度 74/100