Skip to main content
Back to Explore

Qwen3 14B (Non-reasoning)

Alibaba·Released 2025-04-28
Open Source14BApache 2.0

Pricing

Input

$0.23

per 1M tokens

Output

$0.82

per 1M tokens

Blended

$0.38

per 1M tokens

Cheaper than 57% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.38

Monthly

$11.43

vs. Similar Models

OpenAI: GPT-4Q:0.0
$37.50+9743%
Z.ai: GLM 4.5VQ:0.0
$0.90+136%
Google: Gemini 2.5 Flash LiteQ:-0.1
$0.17-54%
Hermes 4 - Llama-3.1 70B (Non-reasoning)Q:-0.1
$0.20-48%

Performance

63

tokens/sec

Faster than 30% of models

1.07

seconds

Faster than 52% of models

1.07

seconds

Faster than 67% of models

Market Median

94 tok/s

33% slower

Median TTFT

1.11s

4% faster

Throughput/Dollar

167

tok/s per $/1M

Speed Comparison

Qwen3 235B A22B 2507 Instruct
63 tok/s+0%
Qwen3 14B (Reasoning)
63 tok/s-1%
Devstral Small (Jul '25)
64 tok/s+1%

Benchmarks

MMLU-Pro
67.5%
GPQA Diamond
47.0%
HLE
4.2%
LiveCodeBench
28.0%
SciCode
26.5%
TerminalBench Hard
5.3%
MATH-500
87.1%
AIME
28.0%
AIME 2025
58.0%
IFBench
23.9%
Long Context Recall
0.0%
Tau2
32.2%
Market AverageTop Score
apache-2.014BGGUF / GPTQ / AWQ
Downloads

698.0K

Likes

12

VRAM (FP16)

16-24 GB

GPU

RTX 4090 / M2 Max

Quick Compare

Similar Models

Compare all 7 models