Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
AdinaYΒ 
posted an update 2 days ago
Post
3391
2025.1 - DeepSeek entered the scene, backed by High Flyer Quant
2026.1 - IQuest enters the game, backed by Uniquant Quant πŸ“ˆ and launching IQuest-Coder on huggingface
https://huggingface.co/collections/IQuestLab/iquest-coder

✨ 40B models: Instruct / Thinking / Loop
✨ Loop = MoE-level performance with only ~5% extra training cost
✨ Native 128K context

model seems pretty aight for its size definitely benchmaxxed though, but the real issue is the horrendous output speed

In this post