SiMa.ai Palette 3.0 Runs Multiple AI Models on One Chip, Boosts Throughput Up to 3.5x
SiMa.ai’s Palette 3.0 runs many AI models on one chip and lifts throughput up to 3.5 times
SiMa.ai released Palette 3.0, a software suite that runs multiple AI models concurrently on its Modalix MLSoC and lifts throughput up to 3.5x. The release adds speculative decoding via EAGLE3, DFlash and Gemma 4 MTP, plus support for large MoE models such as GPT-OSS-20B on the 16GB system-on-module. Open-source code is available on GitHub.
Source:SiliconANGLE · siliconangle.com