ModelsModel releaseLarge Language Models2 sources · Oct 16, 2026

tencent/Hy-Memory-30B-A3B

tencent published the model Hy-Memory-30B-A3B on Hugging Face.

ProofWeights ↗1 community thread

Key points

  • interesting focus: "A 3B active model beats Opus 5 at memory" (whatever that means :D)

Sources (2)

Extractive summary: sentences quoted from the sources.

Hype check

Independently evaluated3 independent looks since

  1. Evaluation · Oct 15, 2026Tencent (Hy4 Preview?) getting in on the rogue agent game
  2. Evaluation · Oct 15, 2026Lethe: Link Inference Attacks For Evaluation of Edge Unlearning Methods
  3. Evaluation · Oct 15, 2026GraphBEAT: Graph-based Bucketized Event Aggregation Over Time for Early Risk Stratification of Gastrointestinal Cancers
  4. Follow-up · Oct 15, 2026Qwen-Audio-Jev: A Unified Interface for Audio Decisions
  5. Follow-up · Oct 15, 2026New H3 Turbo Neat lora is good
  6. Follow-up · Oct 15, 2026How artificial neural networks build their own learning machinery [R]

Before this

  1. Oct 15, 2026StepFun Releases Step 5 Preview Open Weights: 600B MoE, 27B Active, 1M Context
  2. Oct 15, 2026Welcome FLUX.3 Image Klein in Diffusers
  3. Oct 14, 2026ibm-granite/granite-switch-4.1-30b
  4. Oct 14, 2026microsoft/UserRepGen-4B
  5. Oct 14, 2026Introducing Timestral
  6. Oct 13, 2026Run fast, private agents on-device with LFM2.5-24B-A2B

Related