phi-3-mini-4k-instruct-uc-v3-bpw5
Quick start โ try Sipsa near-lossless 5-bit in 30 seconds
pip install ultracompress
uc try sipsa-qwen3-0.6b # 30s demo (free, no signup, no GPU)
This calls Sipsa's free inference API on a compressed model โ see what 1.004x PPL ratio looks like in practice.
To run this model via the API: uc try sipsa-phi-3-mini-4k
Browse the full catalog of try-able models: uc catalog
Need this model in production?
Phase 0 POC: $5K / 5 business days / SHA-256 reproducible-reconstruction audit (deterministic decode to the validated artifact).
Contact: founder@sipsalabs.com | sipsalabs.com/poc
Near-lossless 5-bit compression (~1% perplexity; lossy) of microsoft/Phi-3-mini-4k-instruct โ produced by Sipsa Labs (UltraCompress).
- Reproducible, cryptographically verifiable reconstruction: a deterministic decode to the SHA-256-pinned validated artifact (not bit-identical to the original bf16 model).
- Independently perplexity-verified end-to-end.
- License: BUSL-1.1 + Additional Use Grant.
pip install ultracompress
Documentation & verification: https://sipsalabs.com ยท https://github.com/sipsalabs/ultracompress
Patents pending.
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support
Model tree for SipsaLabs/phi-3-mini-4k-instruct-uc-v3-bpw5
Base model
microsoft/Phi-3-mini-4k-instruct