跳轉到主要內容
AITier

Qwen3.5 27B (Reasoning)

AlibabaQwen開源權重Apache 2.0 · 商用許可

描述

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.

發布日期
2026-02-24
參數規模
27.0B
上下文長度
262K
支援模態
audio, image, text, video

能力雷達圖

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

排行榜排名

領域#排名分數來源
智慧體能力模型榜57
48.0
LS
程式碼能力榜189
69.0
AA
通用能力榜122
64.0
AA
多模態榜74
53.0
LS
科學能力169
63.0
AA

基準測試分數 (LLM Stats)

(LLM Stats (zeroeval))

Agents

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

Chat

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

Code

FullStackBench en60.1%自報
FullStackBench zh57.4%自報

General

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

Healthcare

MedXpertQA62.4%自報
PMC-VQA62.4%自報

Instruction Following

IFBench76.5%自報

Language

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

Long Context

LongBench v260.6%自報

Math

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

Multimodal

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

Reasoning

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

Search

WideSearch61.1%自報

Tool Calling

BFCL-V468.5%自報

Video

MLVU85.9%自報

Vision

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

AA 評測指數

(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 分類評分

(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

定價

輸入價格$0.3 / 1M tokens
輸出價格$2.4 / 1M tokens
混合價格(3:1)$0.825 / 1M tokens

速度

Tokens/秒0.0
首Token延遲0.00s
首回答延遲0.00s

供應商價格排行

供應商價格排行

4 個供應商

最便宜: DeepInfra最貴: Alibaba
供應商輸入輸出
1DeepInfra最便宜
$0
$0
2Novita
$0
$0
3EmpirioLabs AI
$0.086
$0.688
4Alibaba主要
$0.3
$2.4

比較該模型在不同 API 供應商之間的定價。

外部連結

LLM StatsArtificial Analysis