Qwen

Qwen3.6 35B A3B

by Qwentext+image+video->text10 endpoints
Neura Intelligence Index
50.0/ 100
Rank
#180
Confidence
high
Benchmarks
9(23% coverage)

Overview

Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token. It uses a hybrid sparse mixture-of-experts architecture combining Gated...

Capabilities

Text generation
Image understanding
Video understanding
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
TextImageVideo
Output
Text

Technical Specifications

Context Window
262.1K tokens
Max Output
235.9K tokens
Tokenizer
Qwen
Uptime (24h)
9994.9%
Parameters
35000000000B
License
apache_2_0Open Weights
Country
🇨🇳CN
Release Date
April 16, 20265mo

Supported Parameters

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Benchmark Performance

coding

BenchmarkScoreSource
SWE-bench Verified
73.4
verified
SWE-Bench Pro
49.5
verified

reasoning

BenchmarkScoreSource
GPQA Diamond
86
verified
Humanity's Last Exam
21.4
verified

multimodal

BenchmarkScoreSource
MMMU
81.7
verified
CharXiv Reasoning
78
verified
MMMU-Pro
75.3
verified

agent

BenchmarkScoreSource
MCP Atlas
62.8
verified
Toolathlon
26.9
verified

Performance Over Time

Pricing

Per 1M tokens
Input$0.10
Output$0.90
Cache read$0.05
Added
April 27, 2026
Last synced 9/20/2026