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

32
general
40
coding
86
reasoning
57
science
60
agents
80
multimodal

Rankings

Domain#RankScoreSource
Agentic Capability105
38.0
LS
Code Ranking131
68.0
AA
General Ranking92
73.0
AA
Multimodal Ranking36
54.0
LS
Science122
66.0
AA

Benchmark Scores (LLM Stats)

(LLM Stats (zeroeval))

3d

SUNRGBD0.35 / 100SR
Hypersim0.13 / 100SR

Agents

t2-bench79.0%SR
BFCL-V468.5%SR
AndroidWorld_SR64.2%SR
WideSearch61.1%SR
BrowseCompOpenAI (2025)61.0%SR
FullStackBench en60.1%SR
TIR-Bench59.8%SR
FullStackBench zh57.4%SR
OSWorld-Verified56.2%SR
VITA-Bench41.9%SR
Terminal-Bench 2.0Stanford × Laude Institute (2026)41.6%SR
DeepPlanning22.6%SR

Biology

GPQANYU + Cohere + Anthropic (2023)85.5%SR

Chemistry

SuperGPQA65.6%SR

Code

SWE-Bench Verified72.4%SR

Communication

Multi-Challenge60.8%SR

Embodied

EmbSpatialBench0.84 / 100SR

Finance

MMLU-Pro86.1%SR
MMLU-ProX82.2%SR

General

IFEvalGoogle Research (2023)95.0%SR
MMLU-Redux93.2%SR
C-Eval90.5%SR
MAXIFE88.0%SR
Global PIQA87.5%SR
MMMLU85.9%SR
MMMU82.3%SR
Include81.6%SR
MMStar81.0%SR
LiveCodeBench v680.7%SR
IFBench76.5%SR
MMMU-Pro75.0%SR
LongBench v260.6%SR
NOVA-6358.1%SR
SimpleVQA0.56 / 100SR

Grounding

RefCOCO-avg0.91 / 100SR
ScreenSpot Pro70.3%SR
RefSpatialBench0.68 / 100SR

Healthcare

VideoMMMU82.3%SR
SlakeVQA80.0%SR
MedXpertQA62.4%SR
PMC-VQA62.4%SR

Image To Text

OCRBench89.4%SR

Language

LingoQA82.0%SR
WMT24++77.6%SR

Long Context

MLVU85.9%SR
LVBench73.6%SR
AA-LCR66.1%SR
MMLongBench-Doc0.60 / 100SR

Math

HMMT 202592.0%SR
HMMT2589.8%SR
MathVista-Mini87.8%SR
DynaMath87.7%SR
MathVision86.0%SR
CodeForces0.81 / 3000SR
PolyMATH71.2%SR
Humanity's Last Exam48.5%SR

Multimodal

VLMsAreBlind96.9%SR
V*93.7%SR
AI2D92.9%SR
MMBench-V1.192.6%SR
OmniDocBench 1.588.9%SR
VideoMME w sub.87.0%SR
VideoMME w/o sub.82.8%SR
CC-OCR81.0%SR
CharXiv-R79.5%SR
MVBench74.6%SR
MMVU73.3%SR
BabyVision44.6%SR
ZEROBench-Sub0.36 / 100SR
Nuscene15.2%SR
ZEROBench0.10 / 100SR

Reasoning

CountBench0.98 / 100SR
Hallusion Bench70.0%SR
BrowseComp-zh62.1%SR
ERQA60.5%SR
Seal-047.2%SR
OJBench40.1%SR

Spatial Reasoning

RealWorldQA83.7%SR

Vision

ODinW41.1%SR

AA Evaluation Indices

(Artificial Analysis)
Intelligence Index(Artificial Analysis)
34.6
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.9
Gpqa(NYU + Cohere + Anthropic (2023))
0.9
Ifbench(Google Research (2023))
0.8
Lcr(Artificial Analysis)
0.7
Scicode(UIUC + Argonne National Lab (2024))
0.4
Terminalbench Hard(Stanford × Laude Institute (2026))
0.3
Hle(Center for AI Safety + Scale AI (2025))
0.2

LLM Stats Category Scores

(LLM Stats (zeroeval))
Instruction Following
90
Biology
90
Legal
80
Math
80
Physics
80
Structured Output
80
Image To Text
80
Language
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: NovitaMost Expensive: Alibaba
ProviderInputOutput
1NovitaCheapest
$0
$0
2EmpirioLabs AI
$0.086
$0.688
3NanoGPT
$0.27
$2.16
4AlibabaPRIMARY
$0.3
$2.4

Compare pricing across different API providers for this model.

External Sources

LLM StatsArtificial Analysis