models · qwen

QwQ-32B

Qwen QwQ — reasoning-focused model that "thinks before answering." Strong math and science.

What this registry has checked

Quarantine. We hold the publisher’s own published hash for these weights and have recorded it against this entry. We have not downloaded and hashed the weights ourselves, so the hash is the publisher’s claim, not our measurement.

Weight file SHA-256	3fc33cc56d949e466cd9f877209c2001805c7f7a6295b36f43c1858d97a0e986
Method	hf-published-sha256

Re-check this yourself → Source → Full record →

What is recorded

Publisher	qwen
Origin	China
Licence	Apache-2.0
Parameters	32,763,876,352 (32.8B)
Architecture	qwen2
Layers	64
Hidden size	5120
Attention	GQA (40 heads, 8 KV)
Native precision	BF16
Pulls recorded here	88,021
Downloads reported upstream	75,192

Reported scores

MMLU-Pro	76.4
GPQA Diamond	59.3
AIME 2025	29
MATH-500	95.7
LiveCodeBench	63.1
IFBench	38.8
Humanity's Last Exam	7.3

Reported by a third party and recorded here, not re-run by us. Source: artificialanalysis.ai/api/v2.

Measured on real hardware

None of these runs are ours. Each figure is the reporter’s, linked to their own evidence, under their own trust label. The number shown is single-stream decode — one user, concurrency 1 — because aggregate throughput across many users is a different quantity and reads as far faster than anyone will see.

Measured by local.ai

Hardware	Build measured	Decode @ 8K	Memory @ 8K	Decode mode
NVIDIA DGX Spark 128GB	—	3.5 tok/s	67.8 GB	not stated

Measured by local.ai, not by us: 1 run. Each names its engine pinned by image digest and the exact serve command, in the full record.

Elsewhere in the registry

All models → Browse with filters → What will run on my hardware? →

Registry record as of 2026-06-11. Identity, licence and parameter count are properties of a release and do not move; tier can, and the verify link above re-checks it against the live registry rather than this page.