You're saying that this can extend a regular 8k context model up to the level of Gemini 2.0 Pro? What about Qwen 2.5 1M? How far can we extend THAT model's context?
You're saying that this can extend a regular 8k context model up to the level of Gemini 2.0 Pro?
What about Qwen 2.5 1M? How far can we extend THAT model's context?