Qwen3.5 2B (Reasoning)

Qwen3.5 2B (Reasoning)

Alibaba · Released Mar 2, 2026 Professional
Intelligence #10 / 576
82.0 Our Score
AA Index #212 / 378
16.3 Artificial Analysis
Input #133 / 577
$0.020 per 1M tokens
Output #140 / 577
$0.100 per 1M tokens
Context
— Not reported

Analysis Summary

Qwen3.5 2B (Reasoning) is a 2-billion parameter model from Alibaba, priced at $0.02 per million input tokens. The reasoning mode gives it a modest intelligence uplift over the 0.8B variants, with a GPQA score of 0.456 suggesting some capacity for structured reasoning. However, coding capability is very limited and the agentic index is low, ruling it out for technical or multi-step workflows.

For businesses, this model fits cost-sensitive summarisation, basic question answering, or lightweight content generation where a larger model would be economically unjustifiable. Instruction following is moderate, and long-context performance is limited, so it works best on short, well-defined inputs.

At this price point it is a reasonable choice for high-volume, low-stakes automation. Teams needing reliable instruction following, coding support, or agentic behaviour should move to a mid-tier model in the Qwen3.5 family or beyond.

Assessed June 6, 2026

Editorial notes

Qwen3.5 2B (Reasoning) is a small Alibaba model at $0.02/1M tokens with limited coding and agentic capability, suited only to lightweight reasoning or summarisation tasks at scale.

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

Performance Profile

Intelligence2.3Technical1.8Value7.3Content3.1
Intelligence 2.3/10
Technical 1.8/10
Content 3.1/10
Value 7.3/10

How Qwen3.5 2B (Reasoning) compares

Qwen3.5 2B (Reasoning) ranks #212 of 378 AI models we track for overall intelligence, #289 of 315 for coding, #120 of 289 for agentic tasks. At $0.02 per million input tokens it is cheaper than 77% of comparable models.

Performance Indices

Source: Artificial Analysis

16.3 Intelligence Index
3.5 Coding Index
36.4 Agentic Index

Benchmark Scores

Intelligence

GPQA Diamond 45.6% Graduate-level scientific reasoning
HLE 2.1% Humanity's Last Exam
SciCode 2.8% Scientific computing

Technical

TerminalBench Hard 3.8% Agentic terminal tasks
τ²-Bench 69% Conversational agent benchmark

Content

IFBench 31.5% Instruction following
LCR 23.7% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does Qwen3.5 2B (Reasoning) stack up?

Compare side-by-side with other professional 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 $0.02 $0.000020
Output $0.10 $0.000100

Frequently asked questions about Qwen3.5 2B (Reasoning)

How much does Qwen3.5 2B (Reasoning) cost?

Qwen3.5 2B (Reasoning) costs $0.02 per million input tokens and $0.10 per million output tokens.

Is Qwen3.5 2B (Reasoning) good for coding?

On our coding benchmark index, Qwen3.5 2B (Reasoning) ranks #289 of 315 models, placing it in the broader range of the field for code generation and debugging.

Who created Qwen3.5 2B (Reasoning)?

Qwen3.5 2B (Reasoning) is developed by Alibaba and was released on March 2, 2026.