Skip to content
The AI news worth your attention
Only How-to#Model release

#Model release

Sep 27

  1. Xiaomi MiMoAI score62

    Xiaomi MiMo Explains Fixing Tool-Call Repetition in MiMo-V2.6 Models

    Xiaomi MiMo reports that tool-call repetition in MiMo-V2.6 reached over 0.05% of responses across agent harnesses, causing stalled agents and wasted context. The team traced the cause to an RL flooding penalty set at 32 calls per turn, which missed smaller excess behavior, and replaced the approach with a specialized teacher distilled via MOPD. Repetition rates for both Pro and Flash dropped substantially, at roughly $90,000 versus an estimated $2.31 million for the alternative fix.

    AIWhy it matters: The post traces an agent failure to a reward blind spot and compares the costs of two fixes, offering a transferable debugging method for RL-trained tool-calling models.

Sep 21

  1. Tencent HunyuanAI score67

    Tencent Hy4 preview compressed to 214 GiB with mixed-precision quantization

    Tencent Hunyuan says it shrank the 770B-parameter Hy4 preview from roughly 1.5TB to 214 GiB while keeping the parameter count unchanged. The quoted Zhihu post by a Tencent Hunyuan quantization team member describes the method: a 1.25-bit sparse ternary encoding, mixed precision across expert layers, and STQ1_0 CUDA kernels in llama.cpp. The author reports nearly unchanged MRCR retrieval and a small decline in math.

    AIWhy it matters: The quoted Zhihu post explains how Hy4 preview's weights were quantized and kept usable at inference, a concrete engineering case for compressing large MoE models.

May 25

  1. MiniMax BlogAI score67

    MiniMax Explains Why Its Model Failed to Output Certain Rare Chinese Tokens

    MiniMax says the M2 series could not generate the rare token "嘉祺" in names like Ma Jiaqi, and its investigation traced the cause to post-training. The company found the token was learned in pretraining, but low coverage of rare tokens in post-training data caused lm_head vectors to drift. Adding synthetic full-vocabulary repetition data restored generation for these tokens and reduced Japanese-to-Russian mixing from 47% to 1%.

    AIWhy it matters: The post traces a specific token failure through tokenizer, embedding, and lm_head checks, showing a reusable way to diagnose post-training generation problems.

That’s everything