Qwen / Qwen3-VL 235B-A22B Instruct
Qwen3-VL 235B-A22B Instruct BF16 quantization
Estimated until measured during engineering
Before → After
Base releaseVerified checkpoint
235.000B
235.000B
quantizationBF16
Quality retention target
Quality retention target
Production artifactArtifact
Lineage and benchmark report
Lineage and benchmark report
GPU targeting
Weights and KV cacheEstimated until measured during engineering
Estimated VRAMMeasured and certified results are delivered with the project.
Runtime packagingtransformers
Transparent pricing
Not currently orderable