Qwen3-8B represents a new tier in the series. With 8.2 billion parameters, this model retains an architecture of 36 layers and 32 attention heads, but introduces several important improvements — it no longer uses tied embeddings, and its context window has been expanded to 40K tokens, enabling excellent capabilities for handling long documents and complex tasks.
The doubling of parameter count compared to the 4B version significantly enhances response quality across all task types, especially in mathematical reasoning, programming, and advanced analysis. The model excels in tasks requiring multi-step reasoning and deep contextual understanding. Built-in support for both *thinking* and *non-thinking* modes allows performance optimization based on task complexity and available processing time, while the *Thinking Budget* mechanism enables fine-grained control over computational intensity for optimal efficiency.
Qwen3-8B is ideal for advanced professional applications such as financial analysis, medical diagnostics, and legal practice. It is well-suited for building intelligent assistants for professionals, automated technical documentation systems, and educational platforms.
| 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 |
1 | $0.38 | 22.66 | 1.223 | Launch | |
40,960.0 |
1 | $0.53 | 77.34 | 2.516 | Launch | |
40,960.0 |
1 | $0.83 | 124.51 | 2.689 | Launch | |
40,960.0 tensor |
2 | $0.97 | 72.83 | 1.830 | Launch | |
40,960.0 |
1 | $1.02 | 145.65 | 2.682 | Launch | |
40,960.0 tensor |
2 | $1.56 | 6.393 | Launch | ||
40,960.0 |
1 | $1.59 | 3.963 | Launch | ||
40,960.0 |
1 | $2.37 | 169.11 | 11.797 | Launch | |
40,960.0 |
1 | $3.83 | 191.66 | 11.786 | Launch | |
40,960.0 |
1 | $4.11 | 14.061 | Launch | ||
40,960.0 tensor |
2 | $4.61 | 24.610 | Launch | ||
40,960.0 |
1 | $4.74 | 21.701 | Launch | ||
40,960.0 tensor |
2 | $9.40 | 44.416 | Launch | ||
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
40,960.0 |
1 | $0.53 | 51.91 | 1.945 | Launch | |
40,960.0 tensor |
2 | $0.57 | 31.88 | 2.891 | Launch | |
40,960.0 |
1 | $0.83 | 83.49 | 2.118 | Launch | |
40,960.0 tensor |
2 | $0.97 | 1.259 | Launch | ||
40,960.0 |
1 | $1.02 | 54.08 | 2.112 | Launch | |
40,960.0 tensor |
2 | $1.56 | 5.822 | Launch | ||
40,960.0 |
1 | $1.59 | 3.392 | Launch | ||
40,960.0 |
1 | $2.37 | 126.85 | 11.227 | Launch | |
40,960.0 |
1 | $3.83 | 11.215 | Launch | ||
40,960.0 |
1 | $4.11 | 13.490 | Launch | ||
40,960.0 tensor |
2 | $4.61 | 24.039 | Launch | ||
40,960.0 |
1 | $4.74 | 21.130 | Launch | ||
40,960.0 tensor |
2 | $9.40 | 43.846 | Launch | ||
| Name | GPU | TPS | Max Concurrency | |||
|---|---|---|---|---|---|---|
40,960.0 tensor |
2 | $0.57 | 15.23 | 1.307 | Launch | |
40,960.0 tensor |
2 | $0.93 | 3.892 | Launch | ||
40,960.0 tensor |
2 | $1.56 | 80.36 | 4.471 | Launch | |
40,960.0 tensor |
2 | $1.56 | 4.471 | Launch | ||
40,960.0 |
1 | $1.59 | 2.036 | Launch | ||
40,960.0 tensor |
4 | $1.82 | 2.065 | Launch | ||
40,960.0 tensor |
2 | $1.92 | 4.225 | Launch | ||
40,960.0 |
1 | $2.37 | 82.02 | 9.870 | Launch | |
40,960.0 |
1 | $3.83 | 9.859 | Launch | ||
40,960.0 |
1 | $4.11 | 12.134 | Launch | ||
40,960.0 tensor |
2 | $4.61 | 22.455 | Launch | ||
40,960.0 |
1 | $4.74 | 19.774 | Launch | ||
40,960.0 tensor |
2 | $9.40 | 42.262 | Launch | ||
Contact our dedicated neural networks support team at support@immers.cloud or send your request to the sales department at sale@immers.cloud.