deepseek-v4.1POST
Modal training library runs 64k-context DeepSeek V4.1 Flash on 64 B300s
Axiom Bio used Modal's training library to run 64k-context DeepSeek V4.1 Flash reinforcement learning on 64 B300 GPUs.
Modal
@modalArchived 10 days agoOriginally posted 2026.10.22
Modal @modalOriginally posted 2026.10.22
Axiom Bio trains large open models to identify molecules from mass spectrometry data, which requires doing long-context RL. With our training library, they had full control to quickly conduct 64k-context DeepSeek V4.1 Flash runs on 64 B300s.
x.com/modal/status/2113019315796705288Archived 2026.10.22 11:59