Skip to main content
Back to Explore

Qwen3.5 4B

Alibaba·Released 2026-02-27
Open Source4BApache 2.0

Pricing

Input

$0.03

per 1M tokens

Output

$0.15

per 1M tokens

Blended

$0.06

per 1M tokens

Cheaper than 88% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.06

Monthly

$1.80

vs. Similar Models

DeepSeek: R1Q:0.0
$1.15+1817%
Gemini 2.5 Flash (Reasoning)Q:0.0
$0.85+1317%
Gemma 4 26B A4B (Non-reasoning)Q:0.0
$0.20+230%
Qwen3.5 9B (Non-reasoning)Q:+0.2
$0.08+33%

Performance

23

tokens/sec

Faster than 0% of models

0.52

seconds

Faster than 86% of models

87.96

seconds

Faster than 1% of models

Market Median

94 tok/s

76% slower

Median TTFT

1.11s

53% faster

Throughput/Dollar

381

tok/s per $/1M

Speed Comparison

ERNIE 4.5 300B A47B
24 tok/s+3%
Gemma 3 12B Instruct
26 tok/s+14%
MoonshotAI: Kimi K2 0711
27 tok/s+17%

Benchmarks

MMLU-ProNot evaluated
GPQA Diamond
77.1%
HLE
7.8%
LiveCodeBenchNot evaluated
SciCode
16.1%
TerminalBench Hard
18.2%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBench
52.0%
Long Context Recall
55.7%
Tau2
92.1%
Market AverageTop Score
apache-2.04B
Downloads

8.8M

Likes

695

VRAM (FP16)

8-16 GB

GPU

RTX 4070 / M2 Pro

Quick Compare

Similar Models

Compare all 7 models