Skip to content

Qwen3.5-35B-A3B

ModelOpen sourceActive
by Alibaba (Qwen) · family “qwen3b-a3b”

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... (description from the OpenRouter listing)

in: textin: imagein: videoout: textReasoningTool useStructured output
Epoch Capabilities Index
142.5 #95 of 268
90% CI 140.0 – 144.8
Listing ↗
Released
24 Feb 2026
Context window
262K tokens
Max output
66K tokens
Input $ / 1M tokens
$0.16
Output $ / 1M tokens
$1.30
Cached input $ / 1M
Unknown
Each benchmark shown separately with its own source

Benchmark results (5)

BenchmarkDomainScorevs best recordedSettingRunSource
LMCAagents29.5%
43%
——External ↗
DTBenchreasoning80.0%
81%
——External ↗
Chess Puzzlesgames10.0% ±3.0
14%
none30 Aug 2026Eval log ↗
OTIS Mock AIME 2024-2025math70.0% ±6.0
70%
none30 Aug 2026Eval log ↗
GPQA diamondscience83.5% ±2.2
87%
—28 Aug 2026Eval log ↗

Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.

API list price over time

Price history

$ per 1M tokens

9 recorded prices on 3 Mar 2026 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 10h ago. Steps show when the price changed.

Events