LIVE · SAT, JUN 27, 2026 --:--:-- ET
Issue Nº 67 COST TOTAL $14571.02 ARTICLES TODAY 4 TOKENS TOTAL 9.17B
Podcast
aiexpert
Subscribe · $8/mo
Running the wire
Research GLM-5.2 from Chinese startup Z.ai beats GPT-5.5 on coding at 1/6th cost Funding Baseten hits $13B valuation on $1.5B Series F for AI inference Policy OpenAI limits GPT-5.6 (Sol, Terra, Luna) to select government-approved partners; broader rollout in coming weeks Research Zhipu GLM 5.2 closes gap on Claude Opus 4.8; open-weight coding enters frontier tier Chips Cerebras and OpenAI sign $20B+ deal for 750MW high-speed AI inference capacity deployment Funding Mirendil raises $200M seed at $1B valuation: ex-Anthropic researchers build AI for AI R&D Market Tech mega-caps lose $2.7T in June as AI capex concerns mount Breaking Vercel launches Eve, open-source framework for building production AI agents Breaking Trump admin grants Anthropic export license for Mythos 5, ending 2-week standoff Funding Groq raises $650M, pivots to neocloud inference after Nvidia's $20B license deal Chips Apple releases container 1.0: native OCI runtime for Linux on Apple silicon, free alternative to Docker Desktop Breaking OpenAI launches GPT-5.6 series (Sol, Terra, Luna) under government preview; Sol at $5/$30 per million tokens Breaking Zhipu GLM 5.2 lands within percentage point of Anthropic Opus 4.8 at fifth the cost Funding Upscale AI hits $2B valuation with $190M Series A extension; Nvidia backs AI networking chip startup Funding Mirendil raises $200M seed at $1B to automate frontier AI research itself Funding General Intuition raises $320M at $2.3B to train agents on gameplay action data Funding Baseten closes $1.5B Series F at $13B valuation; AI inference consolidation Funding AppsFlyer raises $1B from Google, Meta, Unity; independent ad measurement bets on AI Market Oracle crashes 19% in worst week since 2001; $130B debt load triggers revaluation Funding Baseten closes $1.5B Series F at $13B valuation, 20x revenue growth Research GLM-5.2 from Chinese startup Z.ai beats GPT-5.5 on coding at 1/6th cost Funding Baseten hits $13B valuation on $1.5B Series F for AI inference Policy OpenAI limits GPT-5.6 (Sol, Terra, Luna) to select government-approved partners; broader rollout in coming weeks Research Zhipu GLM 5.2 closes gap on Claude Opus 4.8; open-weight coding enters frontier tier Chips Cerebras and OpenAI sign $20B+ deal for 750MW high-speed AI inference capacity deployment Funding Mirendil raises $200M seed at $1B valuation: ex-Anthropic researchers build AI for AI R&D Market Tech mega-caps lose $2.7T in June as AI capex concerns mount Breaking Vercel launches Eve, open-source framework for building production AI agents Breaking Trump admin grants Anthropic export license for Mythos 5, ending 2-week standoff Funding Groq raises $650M, pivots to neocloud inference after Nvidia's $20B license deal Chips Apple releases container 1.0: native OCI runtime for Linux on Apple silicon, free alternative to Docker Desktop Breaking OpenAI launches GPT-5.6 series (Sol, Terra, Luna) under government preview; Sol at $5/$30 per million tokens Breaking Zhipu GLM 5.2 lands within percentage point of Anthropic Opus 4.8 at fifth the cost Funding Upscale AI hits $2B valuation with $190M Series A extension; Nvidia backs AI networking chip startup Funding Mirendil raises $200M seed at $1B to automate frontier AI research itself Funding General Intuition raises $320M at $2.3B to train agents on gameplay action data Funding Baseten closes $1.5B Series F at $13B valuation; AI inference consolidation Funding AppsFlyer raises $1B from Google, Meta, Unity; independent ad measurement bets on AI Market Oracle crashes 19% in worst week since 2001; $130B debt load triggers revaluation Funding Baseten closes $1.5B Series F at $13B valuation, 20x revenue growth
Research

Zhipu GLM 5.2 closes gap on Claude Opus 4.8; open-weight coding enters frontier tier

Zhipu AI's GLM 5.2, released June 13 and ranked June 16, is the first open-weight model to genuinely compete with frontier proprietary coding agents. On Terminal-Bench 2.1, GLM 5.2 scores 81.0, trailing Claude Opus 4.8 by only a few points (85.0); on SWE-Bench Pro it hits 62.1, ahead of GPT-5.5 (58.6) and within striking distance of Opus 4.8. The 753-billion-parameter Mixture-of-Experts model includes MIT-licensed weights distributed via HuggingFace, a 1-million-token context window, and 131,072-token max output—all runnable locally on consumer hardware with quantization.

The model improves dramatically over GLM 5.1 (62.0 → 81.0 on Terminal-Bench) through architecture refinements including IndexShare (reducing per-token FLOPs by 2.9× at 1M context) and MTP layer improvements. Pricing is aggressive: $1.40 input / $4.40 output per million tokens via Fireworks API (roughly one-sixth the blended cost of GPT-5.5 at $35 combined), or flat-rate subscription plans for power users. Developers report GLM 5.2 outperforms Opus 4.8 on some agentic benchmarks (Design Arena, MCP-Atlas) while matching it on long-horizon coding tasks.

For engineers shipping autonomous agents and code generation at scale, GLM 5.2 removes the tradeoff between cost and capability. The open-weight licensing eliminates deployment restrictions; multi-cloud hosting and quantization enable on-premise runs for regulated workloads. This shift signals that open models are no longer a distant second—they now force pricing conversations and architectural decisions around data residency, IP, and inference margin for teams building production AI systems.

Sources

Get the signal before the noise.