Skip to content

Qwen3.5-Flash

ModelActive
by Alibaba (Qwen) · family “qwen3-flash”

The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. Compared to the... (description from the OpenRouter listing)

in: textin: imagein: videoout: textReasoningTool useStructured output
Epoch Capabilities Index
Not scored by Epoch AI
Listing ↗
Released
25 Feb 2026
Context window
1M tokens
Max output
66K tokens
Input $ / 1M tokens
$0.065
Output $ / 1M tokens
$0.26
Cached input $ / 1M
Unknown
Each benchmark shown separately with its own source

Benchmark results (10)

All benchmarks
BenchmarkDomainScorevs best recordedSettingRunSource
LMCAagents29.1%
43%
——External ↗
DTBenchreasoning82.9%
84%
——External ↗
Mystery Game Puzzlesgames20.0% ±4.0
24%
—27 Aug 2026Epoch ↗
FrontierMath-Tiers-1-3-v2-Privatemath18.2% ±2.3
19%
none28 Aug 2026Eval log ↗
Chess Puzzlesgames21.0% ±4.1
29%
—7 Aug 2026Eval log ↗
SimpleQA Verifiedknowledge20.3% ±1.3
27%
—27 Aug 2026Eval log ↗
FrontierMath-Tier-4-2025-07-01-Privatesupersededmath0.0%
0%
—12 May 2026Epoch ↗
FrontierMath-2025-02-28-Privatesupersededmath6.2% ±1.4
12%
—12 May 2026Epoch ↗
OTIS Mock AIME 2024-2025math84.4% ±5.5
84%
—7 Aug 2026Eval log ↗
GPQA diamondscience82.3% ±2.7
86%
—7 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

8 recorded prices on 3 Mar 2026 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 7h ago. Steps show when the price changed.

Events