tencent/Hy-Memory-30B-A3B
main boardmodel
Hy-Memory-30B-A3B is a compact long-term-memory writer model (~3B active parameters) developed by Tencent Hunyuan to execute the write path inside external memory frameworks. Mixture-of-experts, 30B total, 3B active (128 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.
- Org
- tencent
- Released
- 2026-10-15
- Status
- main board
- License
- apache-2.0 (permissive)
- Parameters
- 30.1B total, 3.0B active
- Architecture
- MoE · 128 experts, 8/token · HYV3ForCausalLM
- Context
- 262,144 tokens
- Task
- text-generation · text
- Signal
- 313 HF likes · 4208 downloads · 1 X post(s)
Runs on
real quant files; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~20.0 GB: 24GB GPU (3090 / 4090), 32GB GPU (5090), 24GB GPU (7900 XTX), Mac 36GB, CPU only, 32GB RAM (slow)
- Q8 ~33.8 GB: 48GB GPU (RTX 6000 Ada / A6000), Mac 48GB, Mac 64GB (M1 Max), CPU only, 64GB RAM (slow)
- FP16 ~61.9 GB: 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
- Hugging Face
- MLX: mlx-community/Hy-Memory-30B-A3B-4bit
- MLX: mlx-community/Hy-Memory-30B-A3B-8bit
- quant: tencent/Hy-Memory-30B-A3B-FP8
- GGUF: tencent/Hy-Memory-30B-A3B-GGUF
First published here 2026-10-15 · updated 2026-10-15 · markdown · JSON