vLLM
vLLM across Glyph
- Engineering blogs
- Paydays: Pay schedule
- Problems people post
- Latest releases: 4 releases
- Open source repos
- Official socials
- Standards support: A business software company that supports OpenAI API, OTel and MCP.
vLLM across Glyph: timeline
- OpenAI users keep asking for help (Problems)
- Running DeepSeek locally hits memory and speed limits (Problems)
- OpenAI users keep asking for help (Problems)
- Running LLMs locally hits memory and speed limits (Problems)
- Running Perplexity locally hits memory and speed limits (Problems)
- Running PyTorch locally hits memory and speed limits (Problems)
- Running NVIDIA locally hits memory and speed limits (Problems)
- Migrating vLLM to the LibTorch Stable ABI (Blogs)
- Zero Mismatch, Simple Async RL: REINFORCE Zero, DeepSeek-V4.1 and Kimi K3.1 (Blogs)
- Trouble with vLLM kernel (Problems)
- Fine-tuning vLLM is costly and unpredictable (Problems)
- DeepSeek users keep filing bugs (Problems)
- Trouble with OpenAI calls (Problems)
- Running Qwen locally hits memory and speed limits (Problems)
- Running vLLM locally hits memory and speed limits (Problems)
- SQLite upgrades break working setups (Problems)
- PyTorch lock-in makes switching costly (Problems)
- OpenAI upgrades break working setups (Problems)
- Trouble with vLLM draft (Problems)
- Qwen users keep filing bugs (Problems)
295 events in all: the full list as JSON.