models · qwen

Qwen2.5-Coder-32B-Instruct

Code-specialized 32B model with open-source SOTA HumanEval / code-completion quality.

Compare Verify

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-256d3bbdd1e787d7e36d63498f035d549e7515af8d6dce0385264a4c98509e0d76a
Methodhf-published-sha256

What is recorded

Publisherqwen
OriginRecorded as a covered nation under 10 U.S.C. § 4872(f) — the PRC, Russia, Iran or North Korea. The registry did not record which.
LicenceApache-2.0
Parameters32,763,876,352 (32.8B)
Context window128K
Modalitytext->text
Architectureqwen2
Layers64
Hidden size5120
AttentionGQA (40 heads, 8 KV)
Native precisionBF16
Pulls recorded here480,013
Downloads reported upstream1,484,515

Reported scores

MMLU-Pro63.5
GPQA Diamond41.7
MATH-50076.7
LiveCodeBench29.5
Humanity's Last Exam3.5

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

HardwareBuild measuredDecode @ 8KMemory @ 8KDecode mode
NVIDIA DGX Spark 128GB—3.5 tok/s67.8 GBnot 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

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.