Qwen3-30B-A3B is the first Mixture-of-Experts (MoE) model in the series, featuring a total of 30 billion parameters, with only 3 billion activated per token. The model is built on an architecture with 48 layers, 32 attention heads, and supports a context window of 40K tokens.
The model employs an innovative Mixture of Experts (MoE) architecture with a total of 128 experts, out of which only 8 are activated per token. This allows achieving inference efficiency comparable to compact 3B-parameter models, while the generation quality approaches that of dense 30B-parameter models. The key advantage of the MoE approach is its optimal balance between high performance and energy efficiency. The system dynamically selects the most suitable experts for each specific task, delivering excellent processing quality at significantly lower computational costs compared to dense models of similar capability.
Qwen3-30B-A3B is well-suited for high-load systems requiring a balance between quality and performance: cloud API services, enterprise chatbots, content automation systems, multi-user AI platforms. The model is ideal for companies that need the quality of large models but operate under limited computational budgets.
| Model Name | Context | Type | GPU | Status | Link |
|---|
There are no public endpoints for this model yet.
An endpoint is a server with pre-installed software that automatically downloads neural network weights and launches an OpenAI-compatible API to handle incoming requests.
A private endpoint is created within the user's project, belongs exclusively to the user, and is billed per server configuration in accordance with the infrastructure pricing.
We recommend creating private endpoints when you need to:
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
40,960.0 tensor |
2 | $0.93 | 6.109 | Launch | ||
40,960.0 |
1 | $1.02 | 1.224 | Launch | ||
40,960.0 tensor |
2 | $1.56 | 6.629 | Launch | ||
40,960.0 |
1 | $1.59 | 3.145 | Launch | ||
40,960.0 tensor |
4 | $1.82 | 3.732 | Launch | ||
40,960.0 |
1 | $2.37 | 156.98 | 14.896 | Launch | |
40,960.0 |
1 | $3.83 | 17.48 | 14.879 | Launch | |
40,960.0 |
1 | $4.11 | 18.291 | Launch | ||
40,960.0 tensor |
2 | $4.61 | 33.955 | Launch | ||
40,960.0 |
1 | $4.74 | 29.751 | Launch | ||
40,960.0 tensor |
2 | $9.40 | 63.665 | Launch | ||
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
40,960.0 tensor |
2 | $0.93 | 2.128 | Launch | ||
40,960.0 pipeline |
3 | $1.06 | 1.186 | Launch | ||
40,960.0 tensor |
4 | $1.26 | 4.258 | Launch | ||
40,960.0 tensor |
2 | $1.56 | 147.35 | 2.648 | Launch | |
40,960.0 tensor |
2 | $1.56 | 2.648 | Launch | ||
40,960.0 tensor |
2 | $1.92 | 103.83 | 2.628 | Launch | |
40,960.0 |
1 | $2.37 | 11.109 | Launch | ||
40,960.0 tensor |
2 | $2.93 | 6.470 | Launch | ||
40,960.0 |
1 | $3.83 | 170.18 | 11.092 | Launch | |
40,960.0 |
1 | $4.11 | 14.505 | Launch | ||
40,960.0 tensor |
2 | $4.61 | 29.973 | Launch | ||
40,960.0 |
1 | $4.74 | 25.964 | Launch | ||
40,960.0 tensor |
2 | $9.40 | 59.683 | Launch | ||
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
40,960.0 tensor |
4 | $1.62 | 5.237 | Launch | ||
40,960.0 pipeline |
6 | $1.65 | 3.827 | Launch | ||
40,960.0 |
1 | $2.37 | 122.46 | 3.832 | Launch | |
40,960.0 tensor |
4 | $2.89 | 141.17 | 6.278 | Launch | |
40,960.0 tensor |
4 | $3.01 | 6.278 | Launch | ||
40,960.0 tensor |
4 | $3.60 | 151.97 | 6.238 | Launch | |
40,960.0 |
1 | $3.83 | 3.815 | Launch | ||
40,960.0 |
1 | $4.11 | 7.227 | Launch | ||
40,960.0 pipeline |
3 | $4.34 | 6.642 | Launch | ||
40,960.0 tensor |
2 | $4.61 | 22.864 | Launch | ||
40,960.0 |
1 | $4.74 | 18.687 | Launch | ||
40,960.0 tensor |
4 | $5.74 | 13.922 | Launch | ||
40,960.0 tensor |
2 | $9.40 | 52.574 | Launch | ||
Contact our dedicated neural networks support team at support@immers.cloud or send your request to the sales department at sale@immers.cloud.