The Dawn of EV, page-1692

I haven't tracked this back to source, but a long twitter post by Battery Bulletin had the table below. LMFP batteries with 3,570 cycles. NCM batteries of different densities with typically 2,000 cycles. Blend the two and you would expect the cycle length to be somewhere between 2,000 (high NMC weighting) and 3,570 (high LMFP weighting). The work attributed to a CATL patent was showing that with a LMFP/NMC blend heavily weighted towards LMFP you could get a life cycle higher than either LMFP or NCM independently. The blends achieving this could have as little as 20% NMC.

This would mean you have a LMFP/NMC blend having higher cycle length and higher density than LFP and presumably similar safety attributes. That would make life tough for LFP if (when) commercialised.

The LMFP used was 60% Mn, 40% Fe so its a variant that would add a lot of HPMSM demand if it were to become a mainstream commercial version.