ScalarLM v1.0

Full Changelog: https://github.com/tensorwavecloud/ScalarLM/commits/v1.0

v0.99 - GPT-OSS and Qwen3

Major changes:

• new model support is verified including GPT-OSS and Qwen3 235b 2507
• vLLM is updated to top of tree
• SLURM integration now supports auto resize of Megatron-LM training workers
• Distributed training on hundreds of AMD GPUs is verified
• Multimodal llm is now supported

Release Images:
gdiamos/scalarlm-amd:v0.99
gdiamos/scalarlm-nvidia-12.0:v0.99
gdiamos/scalarlm-nvidia-9.0:v0.99
gdiamos/scalarlm-nvidia-8.6:v0.99
gdiamos/scalarlm-nvidia-8.0:v0.99
gdiamos/scalarlm-nvidia-7.5:v0.99

New Contributors
@BBBlair made their first contribution in supermassive-intelligence/scalarlm#101
@AnaRubioT made their first contribution in supermassive-intelligence/scalarlm#107
@funston refactored the vLLM integration to simplify it and make it easier to pull from vLLM main