BGE-M3
Readymodel path:accounts/fireworks/models/bge-m3
BAAI bge-m3 multilingual embedding model (XLM-RoBERTa-large, 1024-dim dense, 8192 ctx).
On-demand deployments allow you to use BGE-M3 on dedicated GPUs with Fireworks' high-performance serving stack with high reliability and no rate limits.