GPT-OSS / GPT-OSS 20B
MXFP4 quantization for GPT-OSS 20B
The published profile is indicative; engineering measurements are completed for the selected model and hardware.
Before → After
Base releaseVerified checkpoint
21.000B
21.000B
QuantizationMXFP4
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