TorchSpec merges DFlash2 training support with Qwen3-8B recipe

zhyncs42 · x · 2026-08-21

TorchSpec has merged a PR adding DFlash2 training support. The update introduces DFlash2 configurations, dynamic causal convolution, and candidate selection mechanisms. It also exports the public model configuration and checkpoint schemas used by SGLang and vLLM, and adds a Qwen3-8B training recipe with focused tests.

Original post →

More from Infra

Infra channel →