Qwen3-14B
Parameters
14B
Context Length
131K
Modality
Text
Architecture
Dense
KV Cache Rate
192 KB/tok
License
Apache 2.0
Release Date
29 Apr 2025
Knowledge Cutoff
Jan 2025
API Pricing (per 1M)
Input: $0.12 · Output: $0.24
Hardware Sizing
Precision
Context
Workload Memory: 10.52 GB
INT4 · 8K context
Base Weights: 6.52 GB
KV Cache: 1.50 GB
Runtime & Graphs: 2.50 GB
Minimum Required Hardware
1x NVIDIA L4 24GB
10.5 / 21.6 GB usable
1x RTX 3060 12GB
10.5 / 10.8 GB usable
Apple M4 24GB
10.5 / 18.0 GB usable
Determine requirements for your specific workload:
Open in VRAM Calculator
Architecture Diagram
Evaluation Benchmarks
Rank
#205
| Benchmark | Score | Rank |
|---|---|---|
Agentic Index | auto 0.9 | 118 |
Coding Index | 0.14 | 143 |
Intelligence Index | auto 0.06 Standard 0.07 | 287 275 |
Rankings
Overall Rank
#205
Coding Rank
#146
About Qwen3-14B
Qwen3-14B is a dense foundation model by Alibaba Cloud optimized for multi-step reasoning, mathematical problem solving, and MCP-compatible agent workflows. It features togglable thinking modes and an extensible 131K token context window.
Technical Specifications
Attention
Attention Structure
Grouped-Query Attention
Attention Heads
80
Key-Value Heads
8
Attention Head Dimension
128
Position Embedding
ROPE
RoPE Theta
1,000,000
Normalization
Layer Normalization
Activation Function
SwigLU
Dimensions
Hidden Dimension Size
5,120
Number of Layers
48
FFN Intermediate Size (Dense)
17,408
Tokenizer
Vocabulary Size
151,936
About Qwen 3
The Alibaba Qwen 3 model family comprises dense and Mixture-of-Experts (MoE) architectures, with parameter counts from 0.6B to 235B. Key innovations include a hybrid reasoning system, offering 'thinking' and 'non-thinking' modes for adaptive processing, and support for extensive context windows, enhancing efficiency and scalability.