Qwen3.5 4B (Non-reasoning)

Qwen3.5 4B (Non-reasoning)

Alibaba · Released Mar 2, 2026 Efficient
Intelligence #171 / 561
46.5 Our Score
Speed #42 / 261
172.0 tokens / sec
Input #130 / 561
$0.030 per 1M tokens
Output #154 / 561
$0.150 per 1M tokens
Context
Not reported

Analysis Summary

Qwen3.5 4B (Non-reasoning) sits in the Efficient tier on our leaderboard, ranked #171 of 561 published models on overall intelligence. At $0.030 input and $0.150 output per 1M tokens, it is among the most expensive on the market.

Editorial notes

Qwen3.5 4B (Non-reasoning) is a compact, ultra-cheap model from Alibaba with limited reasoning and coding capability, suited only to lightweight classification or routing tasks at minimal cost.

Assessed May 14, 2026

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence4.1Technical3.6Value7.3Content3.5
Intelligence 4.1/10
Technical 3.6/10
Content 3.5/10
Value 7.3/10

Performance Indices

Source: Artificial Analysis

22.6 Intelligence Index
13.7 Coding Index
49.5 Agentic Index

Benchmark Scores

Intelligence

GPQA Diamond 71.2% Graduate-level scientific reasoning
HLE 7.5% Humanity's Last Exam
SciCode 18.3% Scientific computing

Technical

TerminalBench Hard 11.4% Agentic terminal tasks
τ²-Bench 87.7% Conversational agent benchmark

Content

IFBench 33.3% Instruction following
LCR 28.3% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does Qwen3.5 4B (Non-reasoning) stack up?

Compare side-by-side with other efficient models.

Compare Models

Model Information

ProviderAlibaba
Release Date March 2, 2026
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.03 $0.000030
Output $0.15 $0.000150