Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d268Policy30d67, 15 pendingSources1,014

Qwen3.5 4B

AlibabaReleased Mar 2, 2026Current

Score basis

Qwen3.5 4B is behind the leaders in multimodal tasks, long context, instruction following, and coding. Too few results yet to rate reasoning, agentic tasks, safety, math, multilingual tasks, or factuality.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Limited4 capabilities
  1. Multimodal−33.3%4 of 6
  2. Long Context−34.1%2 of 3
  3. Instruction Following−40.9%2 of 3
  4. Coding−45.0%4 of 10
Not rated6 capabilities

Too few results yet to rate Reasoning, Agentic, Safety, Math, Multilingual or Factuality.

Price

$0.03input$0.15outputper million tokens

From Artificial Analysis · All prices

Cheaper than 96% of 331 priced models · 3:1 input-to-output blend, log scale

Evidence

141results on85benchmarks

  • 5 independently verified
  • 35 aggregator
  • 70 vendor-reported
  • 31 cross-referenced

From 12 sources · latest Oct 8, 2026 · How verification works

Research

62 papers reference Qwen3.5 4B

Qwen3.5 4B benchmark results

141 results on 85 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Multimodal

LimitedFull multimodal ranking

33.3% behind the leader4 of 6 ranked benchmarks measured

Show 10 more multimodal resultsHide 10 multimodal results

Long Context

LimitedFull long context ranking

34.1% behind the leader2 of 3 ranked benchmarks measured

Show 2 more long context resultsHide 2 long context results

Instruction Following

LimitedFull instruction following ranking

40.9% behind the leader2 of 3 ranked benchmarks measured

Show 6 more instruction following resultsHide 6 instruction following results

Coding

LimitedFull coding ranking

45.0% behind the leader4 of 10 ranked benchmarks measured

Show 3 more coding resultsHide 3 coding results

Reasoning

Not enough dataFull reasoning ranking

0 of 6 ranked benchmarks measured

Show 3 more reasoning resultsHide 3 reasoning results

Agentic

Not enough dataFull agentic ranking

0 of 7 ranked benchmarks measured

Show 1 more agentic resultHide 1 agentic result

Math

Not enough dataFull math ranking

0 of 5 ranked benchmarks measured

Show 2 more math resultsHide 2 math results

Factuality

Not enough dataFull factuality ranking

0 of 4 ranked benchmarks measured

Show 1 more factuality resultHide 1 factuality result

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 83 more resultsHide 83 results

Qwen3.5 4B: common questions

Who makes Qwen3.5 4B?

Qwen3.5 4B is made by Alibaba.

When was Qwen3.5 4B released?

Qwen3.5 4B was released on Mar 2, 2026, according to Artificial Analysis.

What is Qwen3.5 4B good at?

Qwen3.5 4B is behind the leaders in multimodal tasks, long context, instruction following, and coding. Too few results yet to rate reasoning, agentic tasks, safety, math, multilingual tasks, or factuality.

How much does Qwen3.5 4B cost?

Qwen3.5 4B costs $0.03 per million input tokens and $0.15 per million output tokens, according to Artificial Analysis. At a mix of three input tokens to one output token, it is cheaper than 96% of the 331 priced models we track.

How many benchmarks has Qwen3.5 4B been tested on?

We track 141 results for Qwen3.5 4B on 85 benchmarks from 12 sources, 5 of them independently verified. The latest was recorded on Oct 8, 2026.

About this record

Where Qwen3.5 4B's numbers come from, and every name it appears under.

Tracked since
Apr 27, 2026
Newest source mention
Oct 3, 2026

Where the results come from

Verification: 141 scores · 5 independently verified · 35 aggregator-attributed · 31 vendor cross-reference · 70 vendor-reported. How these tiers are assigned

From 12 sources on 4 sites. Hugging Face supplies 82 of them; the 5 independently verified results come from 2 sites. Bars are coloured by trust tier.

  • huggingface.co82
  • artificialanalysis.ai36
  • api.llm-stats.com19
  • matharena.ai4

Also known as

qwen 3.5 4bqwen3-5-4bqwen3-5-4b (Non-Reasoning)qwen3-5-4b (reasoning)qwen3-5-4b-non-reasoningqwen3.5 4b (non-reasoning)qwen3.5 4b (reasoning)Qwen3.5-4B (4.7B)qwen/qwen3.5-4b