Inspect attached AI model deployment files without executing model code. DEEPBOM analyzes supported TFLite, ONNX, GGUF, SafeTensors, Core ML, and ExecuTorch files in the browser and reports SHA-256, serialized structure, tensor encodings, quantization evidence, and static findings. Ask follow-up questions to inspect a layer, search and page through model evidence, investigate improvement priorities, or compare available static delegate profiles. Inspect encoded activation fusion while keeping runtime fusion unobserved. Explore Model IR tables and visualizations, and export SVG, PNG, Word-ready bundles, CycloneDX 1.7 artifact evidence, or SPDX 2.3 artifact inventories. Use Send PNG to chat to share a selected visualization or Save via ChatGPT to obtain a download link for a generated file when supported by the host. These actions share the generated export, not the model file. Results separate artifact defects, cautions, and evidence gaps, and explain what static analysis cannot establish. Use the local CLI or local MCP for confidential, very large, sharded, or multi-file artifacts. Ask for optional weight distributions, channel statistics, similarity, singular spectra, sparsity or quantization evidence with explicit decoder and budget coverage. Discover workflows for comparison, pruning investigation, BOM verification, metadata and runtime evidence; continue in the relevant deepbom.org workspace for richer interaction, or use the provided local CLI commands. Web continuation requires selecting the model again; model files and results are not transferred by the link.
