Unsloth GLM-5.2 – How to Run Locally

https://unsloth.ai/docs/models/glm-5.2

#HackerNews #Unsloth #GLM-5.2 #AI #Model #Local #Setup #Run #AI #Tools

#source: venturebeat.com #glm 5.2

Z.ai’s open-weights GLM-5.2 beats GPT-5.5 on multiple long-horizon coding benchmarks for 1/6th the cost

https://venturebeat.com/technology/z-ais-open-weights-glm-5-2-beats-gpt-5-5-on-multiple-long-horizon-coding-benchmarks-for-1-6th-the-cost

Is the release of open-weight GLM 5.2 the new "Deepseek Moment"?

I'm amazed how few people are talking about GLM because the latest model absolutely performs on par or better than the biggest closed-weight models in my real world use, for a fraction of the cost. Having exclusively used the GLM 5.x open models for professional dev work since release, actually they always quietly punched above their weights (pun intended) for some time, and 5.2 only cements it as not being a one-off.

#glm #ai

GLM 5.2 thinking in ... Binary? :D

#zai #glm

Offenes China-LLM besser und günstiger als US-Topmodelle? (GLM 5.2)

https://www.youtube.com/watch?v=-uAV5QUh4lM

#glm #opencode #anthropic #openai #gpt #opus

Update, more slides: Run LLMs Locally

I added sandboxing of OpenCode and llama.cpp with nono and Landlock.
And a new slide to describe jailbreaks with DeepInception.

https://codeberg.org/thbley/talks/raw/branch/main/Run_LLMs_Locally_2026_ThomasBley.pdf

#ai #llm #llamacpp #wllama #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode

🧠 Z.ai ha presentato #GLM-5.2, il nuovo modello flagship progettato per attività di lungo orizzonte: coding agent, debugging complesso, ottimizzazione di sistemi e workflow software che richiedono molte interazioni consecutive.

👉 Dettagli e test: https://www.linkedin.com/posts/alessiopomaro_ai-coding-glm-ugcPost-7473608344130207746-SRV2/

___ 
✉️ 𝗦𝗲 𝘃𝘂𝗼𝗶 𝗿𝗶𝗺𝗮𝗻𝗲𝗿𝗲 𝗮𝗴𝗴𝗶𝗼𝗿𝗻𝗮𝘁𝗼/𝗮 𝘀𝘂 𝗾𝘂𝗲𝘀𝘁𝗲 𝘁𝗲𝗺𝗮𝘁𝗶𝗰𝗵𝗲, 𝗶𝘀𝗰𝗿𝗶𝘃𝗶𝘁𝗶 𝗮𝗹𝗹𝗮 𝗺𝗶𝗮 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿: https://bit.ly/newsletter-alessiopomaro

#AI #GenAI #GenerativeAI #IntelligenzaArtificiale #LLM