Topic
Glm 5 2
5 episodes
-
How to Run Open Source AI Models
Sid Saladi argues that frontier AI vendors (Claude, GPT) bundle model, compute, access, and application into one proprietary stack—trapping users in unpredictable pricing and competitive capture. The counter: open-weight models like GLM-5.2, DeepSeek V4, Qwen, and Kimi are now frontier-adjacent in capability (GLM-5.2 beats GPT-5.5 on coding benchmarks, matches Opus 4.8 on others) and cost roughly one-sixth as much. The real problem isn't model quality anymore; it's that companies like Tesla, Uber, and Meta are hemorrhaging money on metered AI because they can't decouple the stack. The guide walks four layers—model, compute, access, harness—and shows how to own each one deliberately instead of letting a vendor own all four by default.
-
Tencent's Hy3 beats GLM 5.2 at half the size | VentureBeat
Tencent’s new Hy3 MoE model (295B total, 21B active) under Apache 2.0 is a production-first release with strong agent/search metrics and dramatically lower serving cost than GLM-5.2, but still trails Zhipu’s coding leader on recent benchmarks. Laura’s excited about the enterprise upside; Harper wants to see independent validation before betting the stack.
-
Snowflake CEO finds GLM 5.2 competitive with Opus 4.7 at a fraction of the cost
Cooper and Miles dig into Snowflake's claim that GLM-5.2 can hang with Claude Opus 4.7 on a real coding benchmark for much less money, and why the interesting part is not 'GLM wins' but 'cheap models are getting close enough that harness quality and retry policy start to matter more than leaderboard prestige.'
-
Agentic Rl
Cameron Wolfe's 'Agentic RL' argues that training LLMs for agentic work requires shifting from single-turn reasoning to multi-turn trajectory optimization, where the harness (tools, environment, memory) becomes part of the RL loop itself. The central claim is that standard post-training methods fail on long-horizon tasks because they don't account for environment state changes across steps, necessitating new rollout infrastructures and stability techniques like PPO over GRPO for variable-length traces.
-
Z.ai Launches GLM 5.2 With a Usable 1M Token Context, Two Thinking Effort Levels, and No Benchmarks at Launch
Z.ai launches GLM-5.2 with a usable 1M-token context window, two thinking-effort levels (High and Max), and same-day availability across all Coding Plan tiers. The 5x jump from GLM-5.1's 200K window lets coding agents hold entire mid-sized repositories in working memory without constant summarization. Setup is a drop-in swap (base URL + model ID) for Claude Code, Cline, and OpenClaw. Critical caveat: Z.ai published zero benchmarks at launch — no SWE-bench, Terminal-Bench, or Code Arena scores. The 744B MoE backbone (40B active params) is unchanged from GLM-5 lineage; all gains are post-training and context engineering.