modelgrep
Q

Qwen: Qwen3.6 35B A3B

qwen/qwen3.6-35b-a3b

73rd smartest of 178ReasoningToolsJSONVision
Use via OpenRouter ↗
Intelligence
31.5
73rd of 178
Design Elo
Speed
177
19th fastest
Latency
248ms
first token
Input price
$0.150
77th cheapest
Context
262K
262K max out

How it compares

Smarter than59%
of all ranked models
Faster than94%
of all ranked models
Cheaper than74%
of all ranked models

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...

Benchmarks

independent · via OpenRouter
Artificial Analysis57th percentile
Intelligence Index
31.5
Coding Index
17.6
Agentic Index
52.5
GPQA Diamond
82%
Humanity's Last Exam
13%
SciCode
1%
Tau²-Bench (agentic)
85%

Providers & pricing (5)

ProviderIn $/MOut $/MUptime
Parasailfp8$0.150$1.0097.9%
AtlasCloudfp8$0.161$0.96598.2%
AkashMLfp8$0.170$1.2095%
SiliconFlowfp8$0.200$1.6094.8%
WandBfp8$0.250$1.25100%

Specifications

Context window262K
Max output262K
Knowledge cutoff
Input modalitiestext, image, video
Output modalitiestext
Prompt caching
Cache read price$0.050/M
ModeratedNo

Qwen3.6 35B A3B FAQ

How much does Qwen3.6 35B A3B cost?

Qwen3.6 35B A3B costs $0.150 per million input tokens and $1.00 per million output tokens via OpenRouter, making it 77th cheapest of 298 paid models.

How smart is Qwen3.6 35B A3B?

Qwen3.6 35B A3B scores 31.5 on the Artificial Analysis Intelligence Index, ranking 73rd of 178 benchmarked models, with a GPQA Diamond score of 82%.

How fast is Qwen3.6 35B A3B?

Qwen3.6 35B A3B generates around 177 tokens per second with 248ms time-to-first-token (p50), the 19th fastest tracked model.

What is Qwen3.6 35B A3B's context window?

Qwen3.6 35B A3B supports a 262K-token context window and can output up to 262K tokens. It accepts text, image, video input.

Compare head-to-head