Summary
The notice applies to serverless, Fast and US-only serverless endpoints while leaving dedicated deployments unaffected. Fireworks lists replacement models for the affected DeepSeek V4, GLM 5.2, Muse Glimmer 30B and Kimi K2.x variants.
Why it matters
Applications using those hosted model identifiers need migration work before the retirement date to avoid inference failures.
Related coverage
- llama.cpp improves parsing for Qwen3-Coder complex types
- pnpm 11.27 adds explicit build approval and trust-policy cleanup
- GitHub Agentic Workflows adds package-aware updates and safer model accounting