Open source · 25 October 2026
Available nowGuide shows how to run Underdog Saluki 27B with 2‑bit quantization
The guide compares Underdog Saluki 27B with ISTA's GSQ‑RCO quantization, examining GGUF headers, memory usage and BFCL tool‑calling performance in stock llama.cpp.
Why it matters
Benchmark tool‑calling latency when using 2‑bit quantized Saluki 27B to set performance expectations for customers.
For product managers
For your product
What does this mean for your product?
Try it on your product More from this issueAll AI news from 25 Oct →
NVIDIA releases open Nemotron 3.5 Bolt E4B 4GB modelAvailable now
Get this every morning, written for your product.
Deltaday reads the whole AI news cycle and tells you what changed, whether it's real yet, and what to do about it.