{"slug": "glm-5-3", "title": "GLM 5.3", "summary": "Zhipu AI released GLM 5.3, a 743B-parameter mixture-of-experts model with ~40B active per token, achieving a Terminal-Bench 3.0 score of 28.3 (up from 4.6 on GLM 5.2) and a DeepSWE v1.1 score of 66.9 (up from 46.2), while introducing mandatory thinking with three effort levels and a 1M native context. The model also demonstrated emergent cybersecurity capabilities, scoring 84.5% on CyberGym and finding 2,436 vulnerabilities across 269 open-source projects, with open weights delayed about two weeks for safety review, expected around August 28, 2026.", "body_md": "# GLM 5.3\n\nMoE workstation**743B total, ~40B active per token (MoE: 256 routed experts, 8 active + 1 shared).** Same base model as GLM 5.2 - every gain comes from scaled-up post-training, not architecture change. Built on the same stack: IndexShare (long-context), SAO (RL for long-horizon tasks), and slime (open-source async RL). Training environments simulate real professional work, some spanning days of engineer effort.\n\n-\n**Context:** 1M native, same as 5.2. -\n**API change:** thinking is now mandatory with three effort levels (low / high / max) - a breaking change for apps that ran with thinking off.\n\n**Coding (vendor-reported).** Terminal-Bench 3.0 28.3 (vs 4.6 on 5.2), DeepSWE v1.1 66.9 (vs 46.2), Agents’ Last Exam (CLI) 28.5 (vs 23.8). On Z.ai’s private Code Bench it scores 31.4% at ~50K output tokens, beating Claude Opus 4.8’s 29.5% at ~120K, but trails Claude Fable 5 (39.5% at max effort). It still trails GPT-5.6 Sol and Fable 5 on the harder public suites.\n\n**Cybersecurity - the emergent capability.** Z.ai added vulnerability-discovery data expecting incremental single-bug gains; the model instead developed multi-step exploit-chain reasoning. CyberGym 84.5% (up from 77.2%, ahead of Claude Mythos 5 at 83.8% and GPT-5.6 Sol at 83.6%), ExploitBench 54.4% (up from 24.4%), ExploitGym 105 tasks in 2 hours / 130 in 6 (vs 29/39 for 5.2). In real-world testing it found 2,436 vulnerabilities across 269 open-source projects - 1,097 critical/high severity, the oldest dating to 1981. 53 CVEs are publicly disclosed; 2,383 sit under embargo via the Security Disclosure Ledger (cvd.z.ai).\n\n**Availability.** Live now via the GLM Coding Plan and ZCode; per-token API access is rolling out in stages. Open weights are held back ~2 weeks for safety review (expected ~Aug 28, 2026) - the first GLM-series release delayed this way. Self-hosting needs ~1.5TB GPU memory at full precision or ~239GB quantized.\n\n**Honest framing.** All benchmark figures are Z.ai’s own harness configurations. Independent verification awaits the open-weight release at the end of August.\n\n- 743.0B\n- 1000k\n- mit\n- Aug 2026\n\n## What people are building with GLM 5.3\n\nReal demos from X\n\n## Scores\n\n## Or run it in the cloud\n\nLive per-provider pricing, throughput and uptime. Click a column to sort.\n\n| Provider | Type | Input $/M | Output $/M | Cache $/M | Tok/s | Latency | Uptime | Value |\n|---|---|---|---|---|---|---|---|---|\n| Sub | - | - | - | - | - | - | $10.00/mo Coding Plan Lite | |\n| Sub | - | - | - | - | - | - | $30.00/mo Coding Plan Pro | |\n| Sub | - | - | - | - | - | - | $80.00/mo Coding Plan Max |\n\nDefault order: throughput among 95%+ uptime providers, then latency; subscriptions last. Sort by any column. Subscription rows show $/mo in the Value column - per-token columns are \"-\". Affiliate links are marked sponsored / nofollow. Confirm current pricing on the provider's site before committing.\n\n[Detailed API pricing page + JSON endpoint →](/models/glm-5-3/pricing)\n\n[See who runs Zhipu AI in production →](/adoption/zhipu)\n\n## Inference cost over time\n\nData accumulates from the first daily sync - longer ranges populate over time. Prices come from OpenRouter snapshots, not a historical API.", "url": "https://wpnews.pro/news/glm-5-3", "canonical_source": "https://tokenstead.ai/models/glm-5-3", "published_at": "2026-08-14 19:45:15+00:00", "updated_at": "2026-08-14 20:20:50.833764+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-research", "ai-safety", "ai-products"], "entities": ["Zhipu AI", "GLM 5.3", "GLM 5.2", "Claude Opus 4.8", "Claude Fable 5", "GPT-5.6 Sol", "Claude Mythos 5", "Z.ai"], "alternates": {"html": "https://wpnews.pro/news/glm-5-3", "markdown": "https://wpnews.pro/news/glm-5-3.md", "text": "https://wpnews.pro/news/glm-5-3.txt", "jsonld": "https://wpnews.pro/news/glm-5-3.jsonld"}}