Skip to main content
AITier

Qwen3.5 27B (Reasoning)

AlibabaQwenOpen WeightApache 2.0 · Commercial OK

Description

Qwen3.5-27B is a multimodal dense foundation model with 27 billion parameters. It combines strong reasoning, coding, multilingual, long-context, and visual understanding performance in a production-friendly open-weight package with a native 262K context window.

Release Date
2026-02-24
Parameters
27.0B
Context Length
262K
Modalities
audio, image, text, video

Capability Radar

23
general
60
coding
86
reasoning
67
science
60
agents
80
multimodal

Rankings

Domain#RankScoreSource
Agentic Capability57
48.0
LS
Code Ranking192
69.0
AA
General Ranking125
64.0
AA
Multimodal Ranking74
53.0
LS
Science171
63.0
AA

Benchmark Scores (LLM Stats)

(LLM Stats (zeroeval))

Agents

t2-bench79.0%SR
VITA-Bench41.9%SR
DeepPlanning22.6%SR

Chat

IFEvalGoogle Research (2023)95.0%SR
Multi-Challenge60.8%SR

Code

FullStackBench en60.1%SR
FullStackBench zh57.4%SR

General

C-Eval90.5%SR
MAXIFE88.0%SR
Include81.6%SR
AndroidWorld_SR64.2%SR
NOVA-6358.1%SR

Healthcare

MedXpertQA62.4%SR
PMC-VQA62.4%SR

Instruction Following

IFBench76.5%SR

Language

MMLU-Redux93.2%SR
MMLU-Pro86.1%SR
MMMLU85.9%SR
MMLU-ProX82.2%SR
WMT24++77.6%SR

Long Context

LongBench v260.6%SR

Math

HMMT 202592.0%SR
HMMT2589.8%SR
MathVista-Mini87.8%SR
DynaMath87.7%SR
MathVision86.0%SR
CodeForces0.81 / 3000SR
PolyMATH71.2%SR

Multimodal

VideoMME w/o sub.82.8%SR
VideoMMMU82.3%SR
MMMU82.3%SR
TIR-Bench59.8%SR
OSWorld-Verified56.2%SR

Reasoning

Global PIQA87.5%SR
GPQANYU + Cohere + Anthropic (2023)85.5%SR
LiveCodeBench v680.7%SR
CharXiv-R79.5%SR
SWE-Bench Verified72.4%SR
AA-LCR66.1%SR
SuperGPQA65.6%SR
BrowseComp-zh62.1%SR
BrowseCompOpenAI (2025)61.0%SR
Humanity's Last Exam48.5%SR
Seal-047.2%SR
Terminal-Bench 2.0Stanford × Laude Institute (2026)41.6%SR
OJBench40.1%SR

Search

WideSearch61.1%SR

Tool Calling

BFCL-V468.5%SR

Video

MLVU85.9%SR

Vision

CountBench0.98 / 100SR
VLMsAreBlind96.9%SR
V*93.7%SR
AI2D92.9%SR
MMBench-V1.192.6%SR
RefCOCO-avg0.91 / 100SR
OCRBench89.4%SR
OmniDocBench 1.588.9%SR
VideoMME w sub.87.0%SR
EmbSpatialBench0.84 / 100SR
RealWorldQA83.7%SR
LingoQA82.0%SR
CC-OCR81.0%SR
MMStar81.0%SR
SlakeVQA80.0%SR
MMMU-Pro75.0%SR
MVBench74.6%SR
LVBench73.6%SR
MMVU73.3%SR
ScreenSpot Pro70.3%SR
Hallusion Bench70.0%SR
RefSpatialBench0.68 / 100SR
ERQA60.5%SR
MMLongBench-Doc0.60 / 100SR
SimpleVQA0.56 / 100SR
BabyVision44.6%SR
ODinW41.1%SR
ZEROBench-Sub0.36 / 100SR
SUNRGBD0.35 / 100SR
Nuscene15.2%SR
Hypersim0.13 / 100SR
ZEROBench0.10 / 100SR

AA Evaluation Indices

(Artificial Analysis)
Tau2(Sierra + U Toronto + Vector Institute (2025))
93.9
Gpqa(NYU + Cohere + Anthropic (2023))
85.8
Lcr(Artificial Analysis)
77.7
Ifbench(Google Research (2023))
75.6
Terminalbench Hard(Stanford × Laude Institute (2026))
32.6
Hle(Center for AI Safety + Scale AI (2025))
23.9
Intelligence Index(Artificial Analysis)
22.9

LLM Stats Category Scores

(LLM Stats (zeroeval))
Instruction Following
90
Biology
90
Chat
80
Image To Text
80
Language
80
Legal
80
Math
80
Physics
80
Structured Output
80
Embodied
80
Finance
80
Grounding
80
Chemistry
80
Text-to-image
80
Video
80
Long Context
70
Multimodal
70
Reasoning
70
Spatial Reasoning
70
Frontend Development
70
General
70
Healthcare
70
Economics
70
Vision
70
Search
60
Agents
60
Code
60
Communication
60
Tool Calling
60
Spatial
20
3d
20

Pricing

Input Price$0.3 / 1M tokens
Output Price$2.4 / 1M tokens
Blended Price (3:1)$0.825 / 1M tokens

Speed

Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s

Provider Price Ranking

Provider Price Ranking

4 providers

Cheapest: DeepInfraMost Expensive: Alibaba
ProviderInputOutput
1DeepInfraCheapest
$0
$0
2Novita
$0
$0
3EmpirioLabs AI
$0.086
$0.688
4AlibabaPRIMARY
$0.3
$2.4

Compare pricing across different API providers for this model.

External Sources

LLM StatsArtificial Analysis