Notes on Qwen-Max-0428
Qwen-Max-0428, an instruction-tuned model larger than Qwen1.5-110B, now ranks top-10 on Arena and outperforms prior Qwen models on MT-Bench.
Excerpt
API DEMO DISCORD
Previously, we opensourced a series of Qwen1.5 model ranging from 0.5 to 110 billion parameters. Now, we release a larger model, Qwen-Max-0428. Qwen-Max-0428 is an instruction-tuned model for chat service. Very recently, it is available via Chatbot Arena and it has now become the top-10 in the leaderboard. Furthermore, our evaluation of MT-Bench also demonstrates that the new model outperforms our previous largest model Qwen1.5-110B-Chat.
Models MT-Bench Arena Qwen1.
Read at source: https://qwenlm.github.io/blog/qwen-max-0428/