Qwen3.5-35B-A3B
by Qwentext+image+video->text7 endpoints
Neura Intelligence Index
64.3/ 100
Rank
#128—
Confidence
highBenchmarks
9(23% coverage)
Overview
The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. Its overall...
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
81.9K tokens
Tokenizer
Qwen3
Uptime (24h)
10000.0%
Parameters
35000000000B
License
apache_2_0Open Weights
Country
🇨🇳CN
Release Date
February 24, 20268mo
Supported Parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Benchmark Performance
general
| Benchmark | Score | Source |
|---|---|---|
| MMMLU | 85.2 | verified |
coding
| Benchmark | Score | Source |
|---|---|---|
| SWE-bench Verified | 69.2 | verified |
reasoning
| Benchmark | Score | Source |
|---|---|---|
| GPQA Diamond | 84.2 | verified |
| Humanity's Last Exam | 47.4 | verified |
multimodal
| Benchmark | Score | Source |
|---|---|---|
| MMMU | 81.4 | verified |
| CharXiv Reasoning | 77.5 | verified |
| MMMU-Pro | 75.1 | verified |
| ScreenSpot Pro | 68.6 | verified |
agent
| Benchmark | Score | Source |
|---|---|---|
| BrowseComp | 61 | verified |
Performance Over Time
Pricing
Per 1M tokens
Input$0.14
Output$1.00
Cache read$0.05
Added
February 25, 2026
Last synced 11/6/2026