Alibaba

Qwen3.8 27B

Model evaluation profile · Source snapshot checked Aug 21, 2026

AI CENTER CONSENSUS

47.9Current rank #21

Evaluation coverage51.0%Evidence currently available

Evidence confidenceLowCompleteness, not model quality

Input price¥3.00Per 1M tokens

Output price¥12.00Per 1M tokens

SOURCE SCORECARD

How this model performs across evaluations.

Summary snapshot

Original benchmark rank, score, and model identifiers stay visible for auditability. Missing evaluations remain visible instead of being treated as zeroes.

Model evaluation summary

The aggregate evaluation is available below. Original benchmark providers are linked here so each result can be checked at its source.

Consensus47.9Coverage51.0%EvidenceLow

Original benchmark sourcesArtificial Analysis Intelligence Index ↗Artificial Analysis Multilingual Index ↗DeepSWE ↗LiveBench ↗Arena leaderboard dataset ↗APEX-Agents ↗TapTap Maker Benchmark ↗Vals Finance Agent ↗

HOW TO READ THIS

Score is only one signal.

Consensus score estimates relative capability across public evaluations. Coverage tells you how much evidence sits behind the placement, while confidence describes evidence quality rather than intelligence.

MODEL SNAPSHOT

Release date2026-08-14

Current rank#21 / 30

Pricing basisCNY / 1M tokens

Benchmark results link directly to the original evaluation providers listed above.