# Qwen3.5 4B API pricing

No price from the maker itself is on record for Qwen3.5 4B. The lowest of 5 providers is EmpirioLabs AI: $0.04 per 1M input tokens and $0.07 per 1M output tokens (as of 2026-11-12). Context window: 262K tokens, up to 32.8K output tokens per request.

- Type: Text
- Context: 262K
- Max output: 32.8K
- Released: 2026-03-02
- Open weights: Yes

## Price by provider (USD per 1M tokens)

| Provider | Model id | Input | Output | Cached | Notes |
| --- | --- | --- | --- | --- | --- |
| [EmpirioLabs AI](https://docs.empiriolabs.ai) | qwen3-5-4b | $0.04 | $0.07 | $0.02 | Lowest |
| [Merge Gateway](https://docs.merge.dev/merge-gateway) | qwen/qwen3.5-4b | $0.04 | $0.07 | — |  |
| [Thinking Machines](https://tinker-docs.thinkingmachines.ai/tinker/compatible-apis/anthropic/) | Qwen/Qwen3.5-4B | $0.33 | $1.005 | $0.066 |  |
| [Thinking Machines](https://tinker-docs.thinkingmachines.ai/tinker/compatible-apis/anthropic/) | Qwen/Qwen3.5-4B:peft:262144 | $0.33 | $1.005 | $0.066 |  |
| [QVAC](https://www.npmjs.com/package/@qvac/ai-sdk-provider) | qwen3.5-4b | $0 | $0 | — |  |
| [SiliconFlow (China)](https://cloud.siliconflow.com/models) | Qwen/Qwen3.5-4B | $0 | $0 | — |  |

Providers are those listed by models.dev. This site has not vetted them one by one; each provider's own page has the binding price.

Prices and specs come from models.dev (MIT licence), updated 2026-11-12. Each provider's name links to its official docs. A provider priced at 0 usually includes the model in a subscription; it is not necessarily a free API.

https://apibook.ai/models/alibaba/qwen3.5-4b/