@0xSUPERGEMMA

SuperGemma Foundation claims to build uncensored open-source AI models for decentralized inference, with a token ($SUPERGEMMA) on Base mainnet (CA: 0x572c4fa77623652411574c51b5ddb7e1b750aba3).

The project positions itself as a foundation supporting open-source AI builders and has announced partnerships and model releases like "SuperDeepseek-V4-Flash" and "SuperGLM-5.2".

However, multiple red flags emerge: the Twitter bio links to a URL flagged by X as potentially malicious/spammy, the account has modest engagement despite being ~4 years old, and there's limited verifiable evidence of actual model releases or technical substance beyond announcements.

Green flags: Clear positioning in decentralized AI / sovereign AI narrative aligned with current market trends · Multiple partnership announcements with other crypto AI projects · Token deployed on Base mainnet with contract address provided

Red flags: Website URL flagged by X as potentially unsafe/spammy/malicious according to X's URL Policy · Account created July 2022 but still only 2.6k followers - suggests limited organic growth over 4 years · Low engagement (avg 40) relative to follower count and bold claims about "strongest opensource AI model" · No verifiable evidence of actual model repositories, GitHub links, or technical documentation in tweets · Vague claims about model superiority without benchmarks or open-source code links