Qwen3 4B
Qwen/Qwen3-4B
Open Source · chat · open-weights
Open
Alert me on changes
Intelligence
Context
41K
Max output
—
Weights
Open
API $/1M
$0.20 / $0.20
Modalities
text
Released
27 Apr 2025
Intelligence Index via Artificial Analysis · 0–100, higher is better
License: apache-2.0 · Qwen/Qwen3-4B
AI summary
● machine-written
Qwen3 4B: Open-weights 4B chat model with 41K context
Qwen3 4B is an open-weights conversational model released in April 2025, available for self-hosting or API access through providers like Fireworks. The 4-billion parameter model supports a 41K token context window and scores in the 17th–25th percentile on standard benchmarks including MMLU Pro (58.6) and GPQA (39.8).
What's new
- Released April 27–30, 2025
- 41K token context window
- Licensed under Apache 2.0
- Available as open weights for local deployment
- API pricing at $0.20/M input and output tokens via Fireworks
Best for
Cost-effective inference for conversational tasksLocal deployment and fine-tuningSmall-footprint applications with modest reasoning requirements