losclouds

AI Models / Compare

QwQ-32B

Open-weight reasoning specialist; best-in-class 32B for math, coding, and STEM.

Creator
Qwen
Lifecycle
Active
Context
131.1K
Max output
32.8K
Released
Mar 5, 2025
Status
unknown
Input
$0.12 / 1M tokens
Output
$0.15 / 1M tokens
Cached read
— / 1M tokens
Cached write
— / 1M tokens
Batch discount
—%
Source
QwQ-32B pricing
Verified
Apr 5, 2026 (High)

Capabilities

Modalities
text→text
Capabilities
reasoningbatchSupportpromptCachingfunctionCallingstructuredOutputs
Strengths
o1-competitive math/reasoning at 32B, Apache 2.0, Very cheap via Groq
Tradeoffs
Thinking tokens add latency, text-only
Official Links
Intro
https://qwenlm.github.io/blog/qwq-32b/
Docs
https://qwenlm.github.io/blog/
Pricing
https://www.alibabacloud.com/help/en/model-studio/model-pricing
Release note
https://qwenlm.github.io/blog/

Benchmark Coverage

BenchmarkVersionScoreDateSourceNotes
AIME 2024202479.5 %Mar 1, 2025Qwen TeamVendor-reported
MATH-500202490.6 %Mar 1, 2025Qwen TeamVendor-reported
GPQA202465.2 %Mar 1, 2025Qwen TeamVendor-reported

Release History

ReleaseAliasLifecycleRelease DateDeprecationShutdownSummary
QwQ-32Bqwq-32bActiveMar 5, 2025——Current published model family snapshot.

Host Coverage

HostTypeContextPricing NoteDifferences
Groqcloud131.1K$0.12/$0.15 per MTok.Fast inference
Alibaba DashScopefirst-party131.1K$0.29/$0.86 per MTok.—
Migration Guidance

Cheapest high-quality reasoning model. Use Qwen3-235B for maximum quality.

Replacement models: qwen3-235b-a22b

Change Events
DateTypeTitleDescriptionSource
Mar 5, 2025family_addedQwQ-32B publishedInitial public model family launch.QwQ-32B release notes

Other models from Qwen

Qwen2.5 72B, Qwen3 235B-A22B, Qwen3 32B