Qwen3.5 122B A10B (Non-reasoning)

AlibabaQwenOpen WeightApache 2.0 · Commercial OK

Description

Qwen3.5-122B-A10B is a multimodal Mixture-of-Experts model with 122 billion total parameters and 10 billion activated parameters. It combines strong reasoning, coding, long-context, and visual understanding performance with production-friendly efficiency and a native 262K context window.

Release Date

2026-02-24

Parameters

122.0B

Context Length

262K

Modalities

image, text, video

Capability Radar

general

coding

reasoning

scienceest.

agents

multimodal

Science uses a reasoning proxy when dedicated science benchmarks are unavailable.

Rankings

Domain	#Rank	Score	Source
Agents & Tools	45	58.0	LS
Code Ranking	116	57.0	AA
General Ranking	121	63.0	AA
Multimodal Ranking	61	70.0	LS
Reasoning	53	68.0	LS
Science	108	59.0	AA

Benchmark Scores (LLM Stats)

3d

SUNRGBD

0.36 / 100SR

Hypersim

0.13 / 100SR

Agents

t2-bench

79.5%SR

BFCL-V4

72.2%SR

AndroidWorld_SR

66.4%SR

BrowseComp

63.8%SR

FullStackBench en

62.6%SR

WideSearch

60.5%SR

FullStackBench zh

58.7%SR

OSWorld-Verified

58.0%SR

TIR-Bench

53.2%SR

Terminal-Bench 2.0

49.4%SR

VITA-Bench

33.6%SR

DeepPlanning

24.1%SR

Biology

GPQA

86.6%SR

Chemistry

SuperGPQA

67.1%SR

Code

SWE-Bench Verified

72.0%SR

Communication

Multi-Challenge

61.5%SR

Embodied

EmbSpatialBench

0.84 / 100SR

Finance

MMLU-Pro

86.7%SR

MMLU-ProX

82.2%SR

General

MMLU-Redux

94.0%SR

IFEval

93.4%SR

C-Eval

91.9%SR

Global PIQA

88.4%SR

MAXIFE

87.9%SR

MMMLU

86.7%SR

MMMU

83.9%SR

MMStar

82.9%SR

Include

82.8%SR

LiveCodeBench v6

78.9%SR

MMMU-Pro

76.9%SR

IFBench

76.1%SR

SimpleVQA

0.62 / 100SR

LongBench v2

60.2%SR

NOVA-63

58.6%SR

Grounding

RefCOCO-avg

0.91 / 100SR

ScreenSpot Pro

70.4%SR

RefSpatialBench

0.69 / 100SR

Healthcare

VideoMMMU

82.0%SR

SlakeVQA

81.6%SR

MedXpertQA

67.3%SR

PMC-VQA

63.3%SR

Image To Text

OCRBench

92.1%SR

Language

LingoQA

80.8%SR

WMT24++

78.3%SR

Long Context

MLVU

87.3%SR

LVBench

74.4%SR

AA-LCR

66.9%SR

MMLongBench-Doc

0.59 / 100SR

Math

HMMT 2025

91.4%SR

HMMT25

90.3%SR

MathVista-Mini

87.4%SR

MathVision

86.2%SR

DynaMath

85.9%SR

CodeForces

0.85 / 3000SR

PolyMATH

68.9%SR

Humanity's Last Exam

47.5%SR

Multimodal

VLMsAreBlind

96.7%SR

AI2D

93.3%SR

93.2%SR

MMBench-V1.1

92.8%SR

OmniDocBench 1.5

89.8%SR

VideoMME w sub.

87.3%SR

VideoMME w/o sub.

83.9%SR

CC-OCR

81.8%SR

CharXiv-R

77.2%SR

MVBench

76.6%SR

MMVU

74.7%SR

BabyVision

40.2%SR

ZEROBench-Sub

0.36 / 100SR

Nuscene

15.4%SR

ZEROBench

0.09 / 100SR

Reasoning

CountBench

0.97 / 100SR

BrowseComp-zh

69.9%SR

Hallusion Bench

67.6%SR

ERQA

62.0%SR

Seal-0

44.1%SR

OJBench

39.5%SR

Spatial Reasoning

RealWorldQA

85.1%SR

Vision

ODinW

44.5%SR

AA Evaluation Indices

Intelligence Index

35.9

Coding Index

31.6

Tau2

0.8

Gpqa

0.8

Lcr

0.6

Ifbench

0.5

Scicode

0.4

Terminalbench Hard

0.3

Hle

0.1

LLM Stats Category Scores

Biology

Structured Output

Text-to-image

Video

Chemistry

Embodied

Finance

General

Grounding

Healthcare

Image To Text

Instruction Following

Language

Legal

Math

Physics

Spatial Reasoning

Vision

Economics

Frontend Development

Long Context

Multimodal

Reasoning

Tool Calling

Agents

Code

Communication

Spatial

Pricing

Input Price$0.4 / 1M tokens

Output Price$3.2 / 1M tokens

Blended Price (3:1)$1.1 / 1M tokens

Speed

Tokens/sec146.3 tokens/s

Time to First Token1.23s

Time to Answer1.23s

Available Providers

(LS internal units)

Provider	Input Price	Output Price
Novita	400K	3.2M

External Sources

LLM Stats