Qwen3.8-Flash-Next: GPU memory and hosting cost
Open weights · 180 billion parameters · Mixture of experts · 262,144-token context · licence other
No cloud sells this model per token: to use it you run it yourself, on GPU instances you rent.
Run it yourself
It needs 450 GB of GPU memory at bf16, the precision it ships in.
| Cloud | Cheapest instance that fits | GPUs | An hour | A month, around the clock |
|---|---|---|---|---|
| Crusoe | a100-sxm.8x | 8 × A100 80 GB | $18.40 | $13,432 |
| AWS | p4de.24xlarge | 8 × A100 80 GB | $27.45 | $20,036 |
| Together AI | hgx-h100.8gpu | 8 × H100 80 GB | $31.92 | $23,302 |
| OCI | BM.GPU.A100-v2.8 | 8 × A100 80 GB | $32.00 | $23,360 |
| Azure | Standard_ND96amsr_A100_v4 | 8 × A100 80 GB | $32.77 | $23,922 |
| Nebius | gpu-h100-sxm.8gpu-128vcpu-1600gb | 8 × H100 80 GB | $36.00 | $26,280 |
| Google Cloud | a2-ultragpu-8g | 8 × A100 80 GB | $40.55 | $29,602 |
| Alibaba Cloud | ecs.gn8v-8x.16xlarge | 8 × GPU H 96 GB | $58.49 | $42,698 |
More from Qwen
- Qwen3.8 27B
- Qwen3 32B
- Qwen3 235B A22B Instruct 2507
- Qwen3.5 397B A17B
- Qwen3 30B A3B Instruct 2507
- Qwen3 235B A22B
Similar size, open weights
- OpenAI gpt-oss 120B
- GLM 5.3 Flash
- DeepSeek V4 Flash
- DeepSeek V4 Flash 0731
- Llama 4 Scout 17B
- MiniMax M2.5
List prices collected 4 November 2026, refreshed every three hours. No negotiated rates or taxes.