Problemsby IbisPyTorch lock-in makes switching costly
Cloud and tool cost2 posts from 2 people2d active+00 posts in the last 7 days, 1 the 7 days before (fading)Unverified
Posts per day
The posts behind it
2, newest firstPyTorch/Helion project creating enormous amount of B200 requests that are affecting PyTorch/PyTorchI am noticing that pytorch/helion project is testing b200 resource in a strange way Please see https://hud.pytorch.org/hud/pytorch/helion/main/1?per_page=50&name_filter=b200 One user pushed some dozens of commits, which failed some B200 tests unrelated to the B200 runner. The...pytorch/pytorch · nWEIdia6 repliesOpen ↗[RFC] Remove torch.compile dependency from Qwen3.8-Flash-Next NVIDIA path## Motivation The NVIDIA implementation of Qwen3.8-Flash-Next (`Qwen4Exp` internally), introduced in #53896, still applies `support_torch_compile` to the main model and MTP classes. CUDA graph capture does not require this model-level AOT compilation, and the model-specific...vllm-project/vllm · gcanlin1 reactions, 1 repliesOpen ↗
Companies and products named
About this problem
Evidence
2 posts from 2 people in 2 places, about 1 a week over 39 days. Mostly on vllm-project/vllm, pytorch/pytorch. Tools named alongside: NVIDIA, vLLM.
Frustration Frustration 0 of 3· Seen on GitHub issues
How it was grouped
Posts that state a pain and match the "vendor-lock" rule about PyTorch. First post Sep 1, 2026, latest Oct 9, 2026. Unverified: fewer than 3 posts, or one person or place only. Method
The same problem elsewhere
Vendor lock-in makes switching tools costlyGitHub lock-in makes switching costlyKubernetes lock-in makes switching costlyClaude Code lock-in makes switching costlyClaude lock-in makes switching costlyNVIDIA lock-in makes switching costlyllama.cpp lock-in makes switching costlyCodex lock-in makes switching costlyAnthropic lock-in makes switching costlypandas lock-in makes switching costlyApache Airflow lock-in makes switching costlyvLLM lock-in makes switching costly
Other PyTorch problems
PyTorch users keep filing bugsTrouble with PyTorch inputTrouble with PyTorch truePyTorch deprecations force unplanned migrationsPyTorch upgrades break working setupsTrouble with PyTorch workerPyTorch builds are slow or flakyTrouble with PyTorch inductorPyTorch GPU access and drivers are painfulTrouble with PyTorch testPyTorch vulnerabilities are hard to triage and patchRunning PyTorch locally hits memory and speed limits
History
steady to fading
PyTorch lock-in makes switching costly (2 posts)