Released 2026-08-28 · general agent
Deployment evidence
Z.ai · 753B checkpoint; official serving guides describe an approximately 743B-744B architecture total · approximately 39B-40B per token active · 1M tokens
Best-fit scenarios
Frontier text-only coding, cybersecurity, tool-use and long-horizon agent workloads that can support an eight-accelerator FP8 worker or a multi-node BF16 deployment.
Not recommended
- Treating the roughly 39B-40B active-parameter figure as the weight-memory footprint; the full expert checkpoint must be loaded
- Attempting the official BF16 checkpoint on a single 8x H200 node
- Operating a qualifying MaaS business above the license's USD 10 billion threshold without the required Z.AI security review
- Leaving maximum reasoning effort enabled for all latency-sensitive traffic without measuring output length and service cost
Complex coding and long-horizon agents
Vendor-statedZ.ai positions GLM-5.3 as a post-trained GLM-5.2 derivative for complex coding and long-horizon agent work. The reported improvements are vendor evaluations, not ChinaAPI reproductions.
Capability sourceCybersecurity analysis
Public evidenceThe official card publishes CyberGym, ExploitGym and ExploitBench results with named harnesses and evaluation settings. They support candidate selection but do not establish production safety or local quality parity.
Capability sourcePrivate million-token codebase analysis
ChinaAPI inferenceThe text-only 1M context and official self-hosting paths make it a candidate for codebases and records that must remain inside the operator's boundary; practical batch size still depends on KV-cache and latency measurements.
Official launch / minimum
Evidence BOfficial single node fp8
vLLM documents the native FP8 checkpoint on one 8x H200 or 8x H20 node. That is the smallest current official vLLM topology in this dossier, not an absolute minimum or a production capacity claim. Primary recipe
200-person team
Evidence BValidated worker benchmark required
For the defined 200-seat profile, start load testing one 8x H200 FP8 worker; use 8x B200 when the full 1M context is part of the target mix. Add a second independent worker only when maintenance or failure recovery is required. No official recipe maps employee count to capacity. Primary recipe
Commercial API
Evidence BOfficial serving recipe
Use independently deployable FP8 workers with admission control and separate ordinary-, long-context- and reasoning-effort SLOs. The vLLM recipe uses 8x H200/H20 for standard serving and 8x B200 with FP8 KV cache for a full-1M starting point; BF16 requires a multi-node design. Primary recipe
Commercial-use check
GLM-5.3 License
MaaS / hosted service: Commercial use is granted, but a licensee or affiliate that operates a Model-as-a-Service business and exceeds USD 10 billion in aggregate revenue over any consecutive 12-month period must pass Z.AI's security review before commercial use.
Attribution: Retain the copyright and license notice in copies or substantial portions of the weights, comply with applicable laws and the license's use restrictions, and note that the weights are provided without warranty.
Known limitations and open questions
- No ChinaAPI hardware reproduction, quality benchmark or like-for-like comparison between the downloadable checkpoint and hosted route
- The full-1M vLLM example on 8x B200 starts at max-num-seqs 32 and explicitly requires tuning to available KV-cache headroom; it is not a concurrency guarantee
- The 753B checkpoint count and the approximately 743B-744B architecture shorthand come from different official reporting layers and should not be treated as a memory estimate
- Thinking is always enabled and defaults to maximum effort unless low or high is passed explicitly
- Vendor benchmark and capability claims have not been reproduced by ChinaAPI