Qwen3.5 2B (Reasoning)
AlibabaQwenOpen WeightApache 2.0 · Commercial OK
Description
Qwen3.5-2B is a 2 billion parameter vision-language model using Gated DeltaNet hybrid architecture with a 3:1 ratio of linear attention to full softmax attention. It supports 262K native context length and features both thinking and non-thinking modes.
Release Date
2026-03-02
Parameters
2.0B
Context Length
—
Modalities
—
Capability Radar
8
general
17
coding
46
reasoning
22
scienceest.
50
agents
30
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Agentic Capability | 96 | 47.0 | LS |
| Code Ranking | 346 | 22.0 | AA |
| General Ranking | 344 | 32.0 | AA |
| Science | 463 | 16.0 | AA |
Benchmark Scores (LLM Stats)
Agents
t2-bench
48.8%SR
BFCL-V4
43.6%SR
Biology
GPQA
51.6%SR
Chemistry
SuperGPQA
37.5%SR
Communication
Multi-Challenge
33.7%SR
Finance
MMLU-Pro
66.5%SR
MMLU-ProX
52.3%SR
General
MMLU-Redux
79.6%SR
IFEval
78.6%SR
C-Eval
73.2%SR
Global PIQA
69.3%SR
MMMLU
63.1%SR
MAXIFE
60.6%SR
Include
55.4%SR
NOVA-63
46.4%SR
IFBench
41.3%SR
LongBench v2
38.7%SR
Language
WMT24++
45.8%SR
Long Context
AA-LCR
25.6%SR
Math
PolyMATH
26.1%SR
AA Evaluation Indices
Coding Index19.7
Intelligence Index10.2
Tau20.7
Gpqa0.5
Ifbench0.3
Terminalbench V2 10.3
Lcr0.2
Tau Banking0.1
Terminalbench Hard0.0
Scicode0.0
Hle0.0
LLM Stats Category Scores
Instruction Following60
Language60
Structured Output60
General60
Legal50
Math50
Physics50
Reasoning50
Finance50
Healthcare50
Agents50
Biology50
Tool Calling50
Chemistry40
Economics40
Long Context30
Multimodal30
Spatial Reasoning30
Communication30
Vision30
Pricing
Input Price$0.02 / 1M tokens
Output Price$0.1 / 1M tokens
Blended Price (3:1)$0.04 / 1M tokens
Speed
Tokens/sec35.6
Time to First Token0.48s
Time to Answer56.73s
Provider Price Ranking
Provider Price Ranking
1 providers
ProviderInputOutput
1AlibabaPRIMARY
$0.02
$0.1
Compare pricing across different API providers for this model.