Qwen: Qwen3 4B (free)
Qwen3 4B (free) is a compact, zero-cost model with tool and function calling support, suitable for very lightweight experimentation. Its benchmark scores are limited and it is not recommended for serious business applications.
Assessment date: April 4, 2026
Our methodology takes into account a range of factors including pricing, functionality, capabilities, benchmark performance, and real-world applicability. Rankings are reviewed and updated regularly as new models are released. Issues with our rankings? Contact us
Performance Profile
Qwen3-4B is a 4 billion parameter dense language model from the Qwen3 series, designed to support both general-purpose and reasoning-intensive tasks. It introduces a dual-mode architecture—thinking and non-thinking—allowing dynamic switching between high-precision logical reasoning and efficient dialogue generation. This makes it well-suited for multi-turn chat, instruction following, and complex agent workflows.
Capabilities
Architecture
| Modality | Text → Text |
| Tokenizer | Qwen3 |
| Instruct Type | qwen3 |
| Parameters | 4B |
Benchmark Scores
Intelligence
Technical
Benchmark data from Artificial Analysis and Hugging Face
How does Qwen: Qwen3 4B (free) stack up?
Compare side-by-side with other emerging models.
Model Information
| OpenRouter ID |
qwen/qwen3-4b:free
|
| Provider | qwen |
| Release Date | April 30, 2025 |
| Context Length | 40,960 tokens |
| Status | Active |
