Intel adds day-zero support for Meta’s Muse Spark Open
The company announced quantization support to run the 816-billion-parameter open-weight model on Xeon processors and its upcoming GPU D-series.
Event / data period:
Published: · Updated:
Intel announced on 15 October that it had day-zero support for running Meta’s open-weight Muse Spark Open model on Xeon processors and its not-yet-released GPU D-series. The company said it used low-bit quantization for CPU and GPU versions and had run the GPU version across four D-series cards.
Intel describes the model as having 816 billion parameters and a one-million-token context window, and says it is aimed at coding and multimodal workflows. The company says GPU D-series is designed with 160 GB of memory per card, but describes the platform as upcoming. Intel did not publish an independent performance evaluation.
Data
Intel adds day-zero support for Meta’s Muse Spark Open
Intel platform support for Meta’s Muse Spark Open
- · Occurred
Announced model and accelerator
INT4 quantization on Xeon and MXFP4 support on upcoming GPU D-series for Muse Spark Open; Intel reported a four-GPU run.
Methodology
Model and hardware specifications are from Intel’s announcement; this is not an independent performance test, and GPU D-series has not yet launched.
Updated:
Subjects and places
Subject proposals
- Science and technologySuggested
- Preserved editorial category/topic; broad candidate onlywww.intel.com ↗
Geographic references
Only source-provided geographic references are shown. Opening an external map is optional. Optional story maps use OpenStreetMap. Opening a map sends your IP address and this site’s address to the map provider; location data remains readable as text without loading the map.
- United States · Place role: Coverage · Place type: Country or area · Location precision: RegionalEvidence
- Explicit preserved editorial country/region; no city or national centroid inferredwww.intel.com ↗