Ever since Google restructured its Gemini plans to a compute-based usage system, it has struggled to provide users of its agentic coding platform, Antigravity, with enough tokens. The Mountain View-based tech giant, last week, announced a shift from the message-based usage to one that measures the number of tokens exhausted. Once the new system was rolled out, users began complaining about hitting rate limits on Antigravity very quickly. Things were so bad that the tech giant had to increase the Antigravity rate limit twice, but it appears even that was less. Now, the company is solving the problem with a new Gemini 3.5 Flash variant.
Gemini 3.5 Flash (Low) Is Here
In a post on X (formerly known as Twitter), Varun Mohan, who oversees Antigravity at Google DeepMind, said, “We heard concerns that Antigravity consumes many tokens for simple tasks now. So, we’re adding Gemini 3.5 Flash (Low) as a way to optimise token usage for these tasks.” The engineer claimed that the new variant consumes fewer tokens while not compromising on performance.
The original Gemini 3.5 Flash is now known as Gemini 3.5 Flash (Medium), and another 3.5 Flash (High) is aimed at more complex tasks. The Low variant is said to generate roughly 45 percent fewer tokens in output than the Medium variant, while outperforming the High variant on software engineering tasks.
Alongside the announcement, Google has also reset the Gemini quota across all plans this week, allowing users to wrap up their projects without hitting the rate limits. This also includes those on the free tier.
In the same thread, a user complained about the rate limit on the image generation model. User @IamEmily2050 said, “I can generate 1000 images on Codex and can only generate 24 on Antigravity on the Ultra plan?” Mohan responded to the feedback and said, “That’s quite low, I agree. Makes sense to increase.” He, however, did not commit to a higher limit.
Notably, the latest improvements for Antigravity arrive after Google increased the platform’s rate limit twice, to 9x of the original limit after the shift to a compute-based usage measurement system.
Source link
#Googles #Gemini #Flash #Antigravity #Users #Maximise #Usage
