MDL
QwQ 32B
Alibaba
Verified against Artificial Analysis · Jul 13, 2026
Capability
64
percentile index
Price /M
$0.66 / $1
in / out
Context
—
Avg score
47.9
9 benchmarks
AA Intelligence
13.4
index
Speed
33 tok/s
TTFT 0.5s
Released
Mar 5, 2025
About
QwQ 32B is a open-weights model in the QwQ family from Alibaba. Benchmarked on 9 evals, averaging 47.9.
Benchmark scores · 9
avg 47.9 — higher is betterThe research behind this model
all papers mentioning it →Qwen3 Technical Report61 cites · May 14, 2025Scalable Token-Level Hallucination Detection in Large Language ModelsMay 12, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed33 tok/s
Latency (TTFT)0.5s
First answer1m 16s
Price in / out$0.66 / $1 per 1M
Blended price$0.745 / 1M
Math index29.0
Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →
History
full ledger →Capability over time
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$1/M