inference-labs-inc/dsperse

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

1327 stars · 14 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

DSperse is a proving-system-agnostic tool for verifiable AI that decomposes ONNX neural network models into circuit-compatible segments and orchestrates compilation, inference, proving, and verification across pluggable zero-knowledge proof backends. It is written in Rust with Python bindings and a CLI, supporting backends like JSTprove and Expander.

Use cases

• generate zero-knowledge proofs of neural network inference
• split an ONNX model into circuit-compatible slices
• verify that an ML model was executed correctly without revealing weights
• run inference on a sliced model with chained segment outputs
• tile large convolutions for zk-circuit execution
• prove ML model execution across pluggable ZK backends

When to choose

• you need verifiable or zero-knowledge machine learning inference on ONNX models
• you want a full slice-compile-run-prove-verify pipeline via CLI or Rust/Python APIs
• you need a proof-system-agnostic zkML framework with pluggable backends

When to avoid

• you just need standard fast inference without cryptographic proofs
• your models are not in ONNX format
• you need a mature production system - the project is relatively early-stage with a non-standard license

Facets

library · maturity active

machine-learning llm-inference cryptography cli compiler machine-learning artificial-intelligence developer-tools rust python cli cross-platform zkml zero-knowledge-proofs onnx model-slicing verifiable-ai proof-systems cryptography linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="inference-labs-inc/dsperse")