Skip to main contentThe Local Bench

Tool use and agents · Salesforce

xLAM-2 8B

Hugging Face: Salesforce/Llama-xLAM-2-8b-fc-r · checked 2026-10-13

Llama 3.1 8B trained on APIGen-MT multi-turn tool data. Non-commercial licence.

Datasheet

Maker
Salesforce
Parameters
8.03 BVerifiedsafetensors.total from the HF API Source
Licence
CC-BY-NC-4.0VerifiedRestricts commercial use or is not declared. Read the licence before you build on it.
Access
Public download
Task (HF pipeline)
text-generation
First published on HF
2025-03-27
Last updated on HF
2025-05-06
Weights in memory
16 GB at 16-bit · 4.9 GB at ~4.85 bitsEstimateParameters × bits ÷ 8. Excludes context cache and runtime overhead; not every model has a 4-bit build.

Benchmark

46.68%
BFCL V4, overall accuracy · #1 of 3 scored in this category

xLAM-2-8b-fc-r (FC) row, rank 34. Source

Used in

The description summarises the model card at https://huggingface.co/Salesforce/Llama-xLAM-2-8b-fc-r. Parameters and licence are read from the Hugging Face API; the weights-in-memory line is our arithmetic, not a measurement.