unsloth/Qwen3-30B-A3B-FP8

Same weights, converted vs Qwen/Qwen3-30B-A3B-FP8. Same weights stored in another dtype (F8_E4M3/F32 → F8_E4M3/BF16).

Tensors changed: Embeddings 0/1, Norms 0/193, MLP 0/2421, Attention 0/384, lm_head 0/1

On the Hub · Downloads on Model Pulse