Moonlight models
2 Moonlight models from Moonshot AI on Hugging Face, from 16.0B to 16.0B parameters, published by Moonshot AI in BF16, with a 8K-token context. The smallest official model, Moonlight-16B-A3B, needs about 35.7 GB of VRAM at its published precision; the cheapest live fit is RTX A6000 at $0.363/hr.
Pick a size
One row per official Moonlight size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| Moonlight-16B-A3B-Instruct | 16.0B | 35.7 GB | 17.8 GB | 8.9 GB | RTX A6000 | $0.363/hr | 0.95 GB |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (2)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Moonlight-16B-A3B | 16.0B | BF16 | 35.7 GB | RTX A6000 | $0.363/hr |
| Moonlight-16B-A3B-Instruct | 16.0B | BF16 | 35.7 GB | RTX A6000 | $0.363/hr |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
More from Moonshot AI
- Kimi K3 (3 models)
- Kimi K2 (5 models)
- Kimi VL (1 models)
- moonshotai (1 models)