$ cat wiki/models/glm-5-3.md
GLM-5.3
Spec
| Attribute | Value |
|---|---|
| Developer | Z.ai (Z.ai / Zhipu AI) |
| Released | 2026-08-14 (GLM Coding Plan only) |
| Announced | 2026-08-14 |
| Context window | unknown |
| Pricing | GLM Coding Plan from $18/month; no per-token row published — Z.ai's API pricing table still lists GLM-5.2 and has no GLM-5.3 entry |
| License | unknown — open weights stated for "roughly two weeks" out, after a safety evaluation; licence name not published |
| Availability | GLM Coding Plan subscribers |
Context window is unknown deliberately. Z.ai states GLM-5.3 reuses | |
| GLM-5.2's base exactly as it was, and that model carries a | |
| 1,000,000-token window — but a shared base is not a published specification, and | |
inferring the row from a sibling is the error this schema's unknown exists to | |
| prevent (source). |
Release Date
2026-08-14, through the GLM Coding Plan (source).
The weights are not out. Z.ai states API access and open weights will arrive in stages after a safety evaluation, roughly two weeks away (source).
That is a change of pattern, not a delay: GLM-5.2 went from subscriber access (2026-06-13) to open weights and standalone API (2026-06-16) in three days. This release separates them by a stated two weeks and attaches a condition — a safety evaluation — to the second half. See Open-Weights Policy Fight.
Benchmarks
Vendor-stated. No harness is reported as published for any of these figures,
and none of the four benchmarks appears in any snapshot under sources/evals/
(source):
| Benchmark | GLM-5.2 | GLM-5.3 |
|---|---|---|
| Terminal-Bench 3.0 | 4.6 | 28.3 |
| DeepSWE v1.1 | unknown | 66.9 |
| Agents' Last Exam (CLI) | unknown | 28.5 |
| CyberGym | unknown | 84.5% |
| Z.ai's internal evaluations report a ~50% improvement in coding capability | ||
| over GLM-5.2, and coverage ranks GLM-5.3 first among open-source models on | ||
| Terminal-Bench 3.0 and Agents' Last Exam | ||
| (source). |
Two figures are worth reading twice:
- Terminal-Bench 3.0: 4.6 → 28.3. A six-fold move, and also under 29% of the tasks. The release material carries the first reading only. Note the benchmark version: Qwen 3.8 27B and DeepSeek V4-Pro-0813 report Terminal-Bench 2.1, a different suite — the numbers are not comparable across the version boundary. See Eval Harness Configuration.
- Agents' Last Exam CLI: 28.5 against GPT-5.6 Sol's 28.6. A gap of one tenth of a point, run by the party that benefits from it being small, with no published harness on either side. Recorded because Z.ai published it, not because it is a measured tie.
On CyberGym, GLM-5.3's 84.5% is reported to slightly surpass Mythos 5 and GPT-5.6 Sol (and Terra, Luna), while gaps remain against closed frontier models on deep exploitation tasks such as ExploitBench (source). The nearest figure this wiki already holds is DeepSeek V4-Pro-0813's vendor-stated 83.3 on the same benchmark — also unharnessed, also first-party.
Use Cases
Coding and long-horizon agentic work, delivered through a subscription coding product rather than a token API — the same channel GLM-5.2 launched into (source).
The cyber-security framing is Z.ai's own and is the sharpest positioning in the release: it leads with CyberGym rather than with a general coding benchmark. See AI-Enabled Cyberattacks.
2026-08-18 — the access model attached to that framing. GLM-5.3 ships alongside Z.ai's "Shield of Open Source" programme: free security audits, automated code-auditing tools via the ZCode platform, and free model usage quotas for the open-source community, with a restricted tier — "Cybersecurity Trusted Access" — under which the model's most sensitive offensive capabilities are reserved exclusively for verified users (source).
Nothing read reconciles that tier with the pending open weights. This page records weights staged for release around 2026-08-28 after a safety evaluation; a verified-access gate on offensive capability is a control that does not obviously survive publication of the weights it gates. No source read states whether the two plans are the same plan, whether the gate applies to the API only, or what verification consists of.
Compared To
- GLM-5.2 — the same base model, unchanged; every difference between these two pages is post-training
- GPT-5.6 Sol (and Terra, Luna) — the named comparison on Agents' Last Exam CLI and CyberGym
- Claude Fable 5 — coverage describes GLM-5.3's coding and agent capability as "approaching" it; no shared benchmark figure was published
- DeepSeek V4-Pro-0813 — the other Chinese-lab agentic release of the same week, and the only other CyberGym figure this wiki holds
- Qwen 3.8 27B — released the same day, and the opposite trade: weights first, subscription never
Sources
- Zhipu AI's answer to Project Glasswing marks shift for Chinese cyber safety — SCMP → (snapshot)
- Zhipu AI releases GLM-5.3, claims it's the strongest open-weights coding model — the-decoder → (snapshot)
- Zhipu releases GLM-5.3 through its coding service, with weights still two weeks away — MLQ News
- GLM-5.3: Post-Training Alone Rebuilt the Coding Ladder — Digital Applied
- GLM-5.3 Launch: Benchmarks, Pricing & Access — explainx.ai
- GLM-5.3 Pricing — emergent.sh
- GLM-5.3: How Chinese labs keep stride with the frontier — Interconnects
Conflicting Reports
- Base parameter count. Coverage of this release describes the shared base as 743B (source); this wiki records 744B total for GLM-5.2 from its own June capture (source). Third-party on both sides, one billion apart, and Z.ai states the base is unchanged — so at most one of the two roundings is right. Neither page was edited on the strength of the other.