step3: GPU memory and hosting cost
Open weights · 320.97 billion parameters · Mixture of experts · licence apache-2.0
No cloud sells this model per token: to use it you run it yourself, on GPU instances you rent.
Run it yourself
It needs 802.4 GB of GPU memory at bf16, the precision it ships in.
| Cloud | Cheapest instance that fits | GPUs | An hour | A month, around the clock |
|---|---|---|---|---|
| AWS | p5en.48xlarge | 8 × H200 141 GB | $63.30 | $46,206 |
| Azure | Standard_ND96isr_MI300X_v5 | 8 × MI300X 192 GB | $68.64 | $50,107 |
| Google Cloud | a3-ultragpu-8g | 8 × H200 141 GB | $84.81 | $61,909 |
More from Other
- dolphin-2.9.1-yi-1.5-34b
- JiRackUltra_1b
- OTel-2.0-LLM-31B-IT
- moondream2
- TinyLlama-1.1B-Chat-v1.0
- MiniCPM5-2B
Similar size, open weights
- Llama 4 Maverick 17B
- Llama 3.1 405B
- Qwen3 235B A22B
- GLM-5.3-Flash
- DeepSeek-V4-Flash-0731
- DeepSeek-V4-Flash
List prices collected 5 October 2026, refreshed every three hours. No negotiated rates or taxes.