Kimi K3: price on every cloud
Open weights · 2780 billion parameters · Mixture of experts · 1,048,576-token context · licence other
Per token, by cloud
| Cloud | Input, per 1M tokens | Output, per 1M tokens | Cheapest region |
|---|---|---|---|
| Azure | $2.50 | $12.50 | australiaeast · 36 regions |
| Telnyx | $2.70 | $13.50 | global |
| DeepInfra | $2.85 | $14.25 | global |
| AWS | $3.00 | $15.00 | af-south-1 · 36 regions |
| Databricks | $3.00 | $15.00 | us-east-1 |
| Fireworks AI | $3.00 | $15.00 | global · 2 regions |
| Nebius | $3.00 | $15.00 | eu-west2 |
| Snowflake | $3.00 | $15.00 | global |
| Together AI | $3.00 | $15.00 | global |
| Alibaba Cloud | — | — | not sold per token |
| Anthropic | — | — | not sold per token |
| Cloudflare | — | — | not sold per token |
| Core42 | — | — | not sold per token |
| Crusoe | — | — | not sold per token |
| Gemini API | — | — | not sold per token |
| Google Cloud | — | — | not sold per token |
| Groq | — | — | not sold per token |
| HF Endpoints | — | — | not sold per token |
| Huawei Cloud | — | — | not sold per token |
| IBM watsonx | — | — | not sold per token |
| Mistral | — | — | not sold per token |
| OCI | — | — | not sold per token |
| OpenAI | — | — | not sold per token |
| xAI | — | — | not sold per token |
Run it yourself
It needs 1951.1 GB of GPU memory, sized from the weight files it publishes.
| Cloud | Cheapest instance that fits | GPUs | An hour | A month, around the clock |
|---|---|---|---|---|
| OCI | BM.GPU.MI355X.8 | 8 × MI355X 288 GB | $68.80 | $50,224 |
| Nebius | gpu-b300-sxm.8gpu-192vcpu-2768gb | 8 × B300 288 GB | $76.00 | $55,480 |
| AWS | 4 × p4de.24xlarge, joined by EFA | 8 × A100 80 GB in each | $109.79 | $80,145 |
| Azure | 4 × Standard_ND96amsr_A100_v4, joined by InfiniBand | 8 × A100 80 GB in each | $131.08 | $95,688 |
| Google Cloud | 2 × a3-ultragpu-8g, joined by GPUDirect RDMA | 8 × H200 141 GB in each | $169.61 | $123,818 |
| Alibaba Cloud | 3 × ecs.gn8v-8x.16xlarge, joined by eRDMA | 8 × GPU H 96 GB in each | $175.47 | $128,095 |
Several servers: the model is sized across identical servers that hold it together, on instance types whose vendor documents a network between servers; the price is all of them. It is shown where no one server holds the model, or where it costs less than the one that does.
At the cheapest per-token price, running it on OCI BM.GPU.MI355X.8 pays off above roughly 10.0 billion tokens a month.
More from Moonshot AI
- Kimi K2.5
- Kimi K2.6
- Kimi K2.7 Code
- Kimi-K2-Instruct
- Kimi-Linear-48B-A3B-Base
- Kimi-Linear-48B-A3B-Instruct
Similar size, open weights
Inside the UAE: sold in-country by Azure, AWS.
List prices collected 21 November 2026, refreshed every three hours. No negotiated rates or taxes.