Engineering ready
GLM-5.3-Flash FP8 Production Package
GLM / GLM-5.3-Flash · FP8 · Engineering status: Engineering ready
Transformation lineage
Base releaseGLM-5.3-Flash
Model engineeringFP8 Quantization, Runtime packaging, Evaluation and benchmarking
Delivery optionsDownload package / Host with FusionPointAI / Download + host
Transparent pricing
- Base model engineeringUSD 95,700.00
- FP8 QuantizationUSD 5,900.00
- Runtime packagingUSD 2,800.00
- Evaluation and benchmarkingUSD 3,600.00
- Download packageUSD 1,500.00
- Managed deployment setupUSD 3,200.00
- One-time engineeringUSD 112,700.00
Monthly hosting
- 6 x nvidia-h100-sxm-80gbUSD 11,160.00
- Monthly hostingUSD 11,160.00
Quantization
SourceSource
TransformationEstimated until measured during engineering
GPU topology
6 × nvidia-h100-sxm-80gb
transformers
Download or managed hosting
Artifact, config, tokenizer, checksums
GPU allocation, runtime, endpoint and monitoring