AIC++★ 1,025⑂ 142 forksMIT
jesdga95/basalt
Qwen3.8-Flash Inference on Blackwell
C++ inference engine running Qwen3.8-Flash-Next locally on NVIDIA Blackwell GPUs with optimized kernels.
// KEY FEATURES
- Executes Qwen3.8-Flash-Next locally on Blackwell architecture
- Implements custom CUDA kernels for attention and MLP layers
- Supports FP8 quantization for reduced memory and latency
New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram