by Alibaba (Qwen Team)

Qwen 3 14B.

text open weights laptop+ 15B params 131K ctx
Cheapest input
$0.12/M
on DeepInfra
Cheapest output
$0.24/M
on DeepInfra
Smallest GPU
1× Nvidia GeForce RTX 3060
$0.038/hr
Description

About Qwen 3 14B.

Qwen3-14B is a dense open-weights text model with 14.8 billion parameters from the Qwen3 family, released under Apache 2.0. A single model can run in a thinking mode for maths, coding and logic, or a non-thinking mode for everyday dialogue. Qwen describes support for over 100 languages and dialects, plus tool calling for agents. Context is 32,768 tokens natively, extendable to 131,072 with YaRN.

Context window

How much it can remember.

131K tokens ≈ 98,304 English words
4K 32K 128K 1M
Capabilities

What it can do.

· Vision input
· Audio input
· Video input
· Function calling
· Tool use
· JSON mode
✓ Streaming
· Fine-tuning