Skip to content
AI Atlas
ModelPreview

QwQ-32B-Preview

Alibaba Groupfamily · Qwen

Open in Graph
data quality56

Updated 5 h ago · first seen 11 Sept 2026

model_01M29A5BMDGM8TMSCA7T6G60XR

Overview

Identity

Identity block not returned by the API for this entity.

Openness

Openness not classified yet — no sourced evidence to place this model in the ontology.

Key facts

Release date

Source:Qwen — official blogT2observed 12 h agomediumLLM-extracted

Status

Source:Qwen — official blogT2observed 12 h agomediumLLM-extracted

Version

Source:Qwen — official blogT2observed 12 h agomediumLLM-extracted

Architecture

Parameters

Source:Qwen — official blogT2observed 12 h agomediumLLM-extracted

Capabilities

Modalities

Modalities
text
Input
text
Output
text

Capabilities

  • Tool calling

    Unavailable

  • Structured output

    Unavailable

  • Reasoning

    Yes

    Artificial Analysis · T2

  • Vision

    Unavailable

  • Audio

    Unavailable

  • Fine-tuning available

    Unavailable

Context window

Source:Artificial AnalysisT2observed 10 h agomedium

Benchmarks3

Compare with another model →

Comparable same task and conditions · Partially comparable same task, conditions differ (effort, temperature, judge) · Not comparable different variant or metric

No benchmark results recorded

Results appear when a tier 1–3 source publishes them; we never copy scores without a source.

Hardware fit37

Estimated

36 of 37 device × quantization combinations fit.

Run locally: your machine →
Estimated hardware fit
HardwareQuantizationDevice memoryEst. memoryFits
Apple M3 Ultra4bit—18.9 GB est.Yes
Mac Studio (Apple M5 Ultra)4bit—18.9 GB est.Yes
Apple M2 Ultra4bit—18.9 GB est.Yes
Apple M1 Ultra4bit—18.9 GB est.Yes
Apple M3 Max4bit—18.9 GB est.Yes
Apple M4 Max4bit—18.9 GB est.Yes
Mac Studio (Apple M5 Max)4bit—18.9 GB est.Yes
MacBook Pro (Apple M5 Max)4bit—18.9 GB est.Yes
Apple M2 Max4bit—18.9 GB est.Yes
Apple M1 Max4bit—18.9 GB est.Yes
Apple M4 Pro4bit—18.9 GB est.Yes
Mac mini (Apple M5 Pro)4bit—18.9 GB est.Yes
MacBook Pro (Apple M5 Pro)4bit—18.9 GB est.Yes
Apple M3 Pro4bit—18.9 GB est.Yes
Apple M1 Pro4bit—18.9 GB est.Yes
Apple M2 Pro4bit—18.9 GB est.Yes
Apple M44bit—18.9 GB est.Yes
iMac (Apple M4)4bit—18.9 GB est.Yes
Mac mini (Apple M6)4bit—18.9 GB est.Yes
MacBook Air (Apple M5)4bit—18.9 GB est.Yes
MacBook Pro (Apple M5)4bit—18.9 GB est.Yes
Apple M24bit—18.9 GB est.Yes
Apple M34bit—18.9 GB est.Yes
NVIDIA GeForce RTX 30904bit24 GB18.9 GB est.Yes
NVIDIA GeForce RTX 40904bit24 GB18.9 GB est.Yes
NVIDIA GeForce RTX 50904bit32 GB18.9 GB est.Yes
NVIDIA A100 80GB4bit80 GB18.9 GB est.Yes
NVIDIA H100 SXM4bit80 GB18.9 GB est.Yes
NVIDIA H100 NVL4bit94 GB18.9 GB est.Yes
NVIDIA DGX Spark4bit128 GB18.9 GB est.Yes
NVIDIA H2004bit141 GB18.9 GB est.Yes
NVIDIA H200 NVL4bit141 GB18.9 GB est.Yes
NVIDIA B2004bit180 GB18.9 GB est.Yes
AMD Instinct MI300X4bit192 GB18.9 GB est.Yes
AMD Instinct MI325X4bit256 GB18.9 GB est.Yes
NVIDIA DGX B2004bit1,440 GB18.9 GB est.Yes
Apple M14bit—18.9 GB est.No
Assumptions (6)
  • Estimated, not measured: weights = parameters × bytes/param × 1.15 runtime overhead.
  • bytes/param: 4bit = 0.5, 8bit = 1.0, fp16 = 2.0 (uniform quantization, no per-layer exceptions).
  • KV cache approximated at 0.5 GB per 8 192 tokens of context, independent of architecture (GQA/MLA models need less).
  • A model 'fits' when the estimate is at most the device memory minus 2 GB reserved for the OS and framework.
  • Mixture-of-experts models are estimated on total parameters (all experts must be resident); active parameters are ignored.
  • Device memory uses the largest configuration when several are listed (e.g. Apple silicon tiers).

Timeline5

Full timeline →

Change history4

Temporal, append-only claims: a new observation closes the previous claim instead of overwriting it. Rewind the record with the as-of picker.

Viewing AI Atlas as of 1 Jan 2025 — attributes exactly as the atlas knew them on that day; later corrections are not shown.

Back to today →

QwQ-32B-Preview was not yet in AI Atlas on 1 Jan 2025

First seen 11 Sept 2026. Nothing is inferred backwards: no attribute is shown for dates before the first observation.
4 claims · 4 properties

Statusstatus1

Claim history for Status
ValueValid from → toStatusSourceConfidenceExtractor
deprecated → currentcurrentArtificial AnalysisT2mediumdeterministic

Opennessopenness1

Claim history for Openness
ValueValid from → toStatusSourceConfidenceExtractor
open-weights → currentcurrentArtificial AnalysisT2mediumdeterministic

Context windowcontext_length1

Claim history for Context window
ValueValid from → toStatusSourceConfidenceExtractor
32.8K tokens → currentcurrentArtificial AnalysisT2mediumdeterministic

Reasoningreasoning1

Claim history for Reasoning
ValueValid from → toStatusSourceConfidenceExtractor
Yes → currentcurrentArtificial AnalysisT2mediumdeterministic

Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →

Provenance

Attributed facts

12

Source tiers

T212

Freshest observation

10 h ago

Conflicts

None

Source documents 2

Source documents
SourceDocumentTypeTierLast observedSnapshots
Qwen — official blogqwenlm.github.io/blog/qwq-32b-preview newsT1· Official12 h ago1
Artificial Analysisartificialanalysis.ai/leaderboards/models leaderboardT2· Quality secondary10 h ago1

Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.

Data quality (56/100) measures how well AI Atlas knows this entity — completeness, primary-source ratio, freshness, conflicts — never how good the model is. Methodology →