Phi-3.5-MoE Instruct vs Skywork-R1V3-38B

Phi-3.5-MoE Instruct

Microsoft

Phi-3.5-MoE-instruct is a sparse mixture-of-experts model from Microsoft's Phi research team, released in August 2024 with 42 billion total parameters across 16 experts and approximately 6.6 billion active parameters per forward pass. The model applies Microsoft's small-data, high-quality training philosophy — developed across earlier Phi generations — to a MoE architecture, targeting reasoning quality comparable to much larger dense models at a fraction of the inference compute. Released under the MIT license, it was notable in the research community for demonstrating that MoE efficiency gains could be realized at smaller total parameter counts than typical large-scale MoE deployments.

Kunlun Tech

Skywork-R1V3-38B, released by Kunlun Tech's Skywork AI team on July 9, 2025, is a 38 billion parameter multimodal reasoning model built on InternVL-38B with reinforcement learning post-training that enhances both visual and textual reasoning. It uses the GRPO algorithm and cold-start fine-tuning to improve reasoning across image and text modalities. Skywork-R1V3-38B targets open-source multimodal reasoning deployments requiring strong performance across vision-language benchmarks.