Intern S2 Mobius Released

·2026.08.05 11:49

Key point

InternLM has released Intern S2 Mobius, a derivative model of Qwen3.5-35B.

Details

InternLM has released Intern S2 Mobius, a derivative model of Qwen3.5-35B, on Hugging Face.

According to the model card, the architecture has been partially modified to apply a design that increases throughput and reduces token usage. However, the performance and efficiency improvement figures require further verification through benchmarks from the official model card and actual serving environments.

llm open-source qwen model-serving internlm

Read originalShare on Threads

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.