Skip to content
llmwaves

Qwen3.5-35B-A3B

Qwen · released Feb 25, 2026

Open weightsReasoningTool useStructured outputVisionVideo in

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...

Specification

Context window
262K
Max output
16K
Knowledge cutoff
Not stated
Parameters
36B
Licence
apache-2.0
Serving providers
7
Moderated
No
Uptime
100.0%

Intelligence

19.3

65th percentile

Coding

—

Coding Index

Agentic

—

Agentic Index

Output speed

121 t/s

Median across providers

Latency

410ms

Time to first token

Cost per task

$0.09

Estimated

Benchmarks

Where the score comes from

The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.

Evaluation scores

Percentage correct · higher is better

  • τ²-bench (Telecom)
    89.2%
  • GPQA Diamond
    84.5%
  • IFBench
    72.5%
  • AA-LCR (long context)
    72.0%
  • Terminal-Bench Hard
    26.5%
  • Humanity's Last Exam
    21.0%

An evaluation missing from this list was not run for this model — it is not a zero.

View as table
EvaluationScore
τ²-bench (Telecom)89.2%
GPQA Diamond84.5%
IFBench72.5%
AA-LCR (long context)72.0%
Terminal-Bench Hard26.5%
Humanity's Last Exam21.0%

Against its peers

Intelligence Index · this model highlighted, nearest peers in grey

Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.

View as table
ModelIntelligence
o320.2
Ling 3.0 Flash20.1
Step 3.7 Flash19.5
Qwen3.5-35B-A3B19.3
Claude Sonnet 4.519.3
LongCat 2.019.1
MiniMax M218.6
Claude Opus 4.118.6

Percentile among all indexed models

Intelligence65th
Terminal work68th

Pricing

What it costs to run

List prices per million tokens, plus what one representative task works out to.

List price

Input / 1M tokens
$0.15
Output / 1M tokens
$1
Cached input / 1M
$0.05
Blended 3:1
$0.362

One task, estimated

$0.09

Input tokens
50,000
Output tokens
80,000
Profile
Reasoning

Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).

Serving providers

Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.