CMSManhattan·released 2026-08-291 source
Not yet benchmarked
Claims drawn from cited facts, not live model generation.
Developed by CMSManhattan and released in 2026, this model carries 27.32 billion parameters, placing it in the large-scale tier rather than the compact end of the spectrum. That scale buys substantially more representational headroom and capability than smaller models, but it comes with the trade-off of heavier compute demands and higher serving costs.
4 cited facts
This model is not yet benchmarked, as it is currently scored on zero tracked benchmarks. It holds no current top score among those benchmarks.
2 cited facts
The average SOTA gap is unknown, so this position cannot be quantified against benchmark leaders. It holds the top score on 0 benchmarks, which means no category leadership yet.
2 cited facts
2 facts cross-checked across data sources: 2 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: huggingface_models
JiRackDeltaNet_27b is an AI model developed by CMSManhattan, released 2026-08-29. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.