ByteDance-Seed/Qwen3-30B-A3B-MoE-SAE
main boardmodel
Feature extraction model from ByteDance-Seed (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 30B total, 3B active. License: apache-2.0.
- Org
- ByteDance-Seed
- Released
- 2026-11-05
- Status
- main board
- License
- apache-2.0 (permissive)
- Parameters
- 30.0B total, 3.0B active
- Architecture
- MoE
- Task
- feature-extraction · embedding
- Signal
- 2 HF likes · 0 downloads
Runs on
estimate from params; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~21.6 GB (estimated size): 24GB GPU (3090 / 4090), 32GB GPU (5090), 24GB GPU (7900 XTX), Mac 36GB, CPU only, 32GB RAM (slow)
- Q8 ~35.4 GB (estimated size): 48GB GPU (RTX 6000 Ada / A6000), Mac 48GB, Mac 64GB (M1 Max), CPU only, 64GB RAM (slow)
- FP16 ~63.5 GB (estimated size): 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), Strix Halo 128GB unified (~96GB usable), Mac 96GB, Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)
Links
First published here 2026-11-05 · updated 2026-11-06 · markdown · JSON