GLM-5.3
- Provider
- Zhipu AI
- Status
- Current
- Context
- 1,000,000 tok
- Price
- $1.4 / $4.4 /MTok
GLM-5.3 is Zhipu AI (Z.ai)‘s current flagship, released on 14 August 2026 through the Z.ai API and the GLM Coding Plan. It is the most unusual release of the year so far for two reasons. First, Z.ai retrained nothing: GLM-5.3 runs on exactly the same base model as GLM-5.2, and every gain comes from scaled-up post-training — Terminal-Bench 3.0 moved from 4.6 to 28.3 and DeepSWE v1.1 from 46.2 to 66.9 on that basis alone (Z.ai docs). Second, and far more consequential for a lab that built its reputation on MIT-licensed weights, Z.ai withheld the weights at launch — the model’s vulnerability-discovery ability grew faster than the company expected as training scaled, and it delayed the open release by roughly two weeks for safety evaluation and hardening.
The capability claim is independently corroborated, which is what earns it a place near the top of our board rather than in the open-weight tier. Artificial Analysis scores GLM-5.3 (max) at 60 on its Intelligence Index — one point behind Grok 4.6 and GPT-5.6 Sol Max (61), three points behind Claude Opus 5 (63), and level with Kimi K3 at a fraction of its price — at $1.40/$4.40 per million tokens, unchanged from GLM-5.2. That combination gives it $0.68 per completed index task, the lowest cost per task of any model in the frontier cluster. The honest caveats: it publishes no SWE-bench figure of any kind, it is text-only where its rivals take images, and on 19 August 2026 the promised weights had not yet shipped.
Quick specs
| Provider | Zhipu AI (Z.ai) |
| Released | 14 August 2026 |
| Status | Available (API and GLM Coding Plan; weights withheld) |
| Architecture | Mixture-of-experts, ~750B total / ~40B active — same base as GLM-5.2 |
| Context window | 1,000,000 tokens |
| Max output | 131,072 tokens |
| Modalities | Text in, text out; multilingual |
| Licence | Not yet published — open weights announced, not shipped |
| Input price | $1.40 / MTok (cached $0.26) |
| Output price | $4.40 / MTok |
| AA Intelligence Index | 60 (max effort, independent) |
| Cost per task | $0.68 — lowest in the frontier cluster |
| Best for | Long-horizon coding agents at frontier quality and open-model prices |
| Limitations | No SWE-bench figure; text-only; weights withheld; China-hosted API |
What GLM-5.3 is
GLM-5.3 is a mixture-of-experts model of roughly 750 billion total parameters with about 40 billion active per token — the same architecture, and the same trained base, as GLM-5.2. Z.ai is explicit that this is a post-training release: no new pretraining run, no new foundation, just far more reinforcement learning on long agentic trajectories. That makes the version-on-version deltas unusually clean, because the only variable that changed is the post-training.
It keeps the 1-million-token context and 128K maximum output of its predecessor, and it is text-in, text-out — there is no vision path, which is the clearest capability gap against GPT-5.6 Sol, Claude Fable 5 and Grok 4.6. One migration detail matters if you are upgrading: reasoning is always on. GLM-5.3 does not support thinking.type: "disabled", so applications that used it must switch to enabled with reasoning_effort: "low" (Z.ai docs). Effort levels are low, high and max, defaulting to max.
Benchmark performance
Two things are worth separating: Z.ai’s own launch table, and Artificial Analysis’s independent run.
Independent (Artificial Analysis)
| Metric | GLM-5.3 (max) | Context |
|---|---|---|
| Intelligence Index | 60 | 1 behind Grok 4.6 and GPT-5.6 Sol Max (61); 3 behind Opus 5 (63); level with Kimi K3 (max) |
| Cost per index task | $0.68 | Lowest in the frontier cluster — under Grok 4.6 and Kimi K3 (both $0.84) |
| Output speed | 84.7 tokens/sec | Above the 74 t/s median, and more than double Kimi K3’s 38.2 t/s |
| Verbosity | 170M output tokens over the eval | Against a 72M median — notably verbose |
Vendor (Z.ai launch table)
| Benchmark | GLM-5.2 | GLM-5.3 | Kimi K3 | Fable 5 | GPT-5.6 Sol |
|---|---|---|---|---|---|
| Terminal-Bench 3.0 | 4.6 | 28.3 | 17.4 | 33.7 | 34.6 |
| DeepSWE v1.1 | 46.2 | 66.9 | 67.5 | 69.7 | 72.7 |
| SWE-Marathon v1.1 | 19.4 | 42.5 | 48.1 | — | — |
| Terminal-Bench 2.1 | 81.0 | 88.2 | 88.3 | — | 88.8 |
| AutomationBench | 26.2% | 48.2% | 46.7% | 46.2% | 45.8% |
| HLE (with tools) | 54.7% | 62.5% | 59.8% | 63.9% | 64.5% |
| Z.ai Code Bench (max) | 23.4% | 34.5% | — | 39.5% | — |
The shape is consistent: GLM-5.3 clears Kimi K3 on the long-horizon terminal work and lands within a few points of the closed frontier, while losing the hardest agentic-coding rows to Fable 5 and Sol. The gap that matters most for our ranking is the one that isn’t there at all — Z.ai published no SWE-bench Pro or Verified score, in any harness, having reported 62.1% on SWE-bench Pro for GLM-5.2 two months earlier.
The cyber capability — and why the weights were held
The most-discussed part of this launch isn’t coding. Z.ai reported that GLM-5.3’s vulnerability-discovery ability kept compounding as post-training scaled, past what the company had planned for, with the model reasoning across full exploitation chains rather than finding isolated bugs. The benchmark movement is the steepest on the card: CyberGym 77.2% to 84.5% and ExploitBench 24.4% to 54.4%, with ExploitGym solves going from 29 to 105 tasks inside a two-hour budget (Z.ai docs). On the cyber rows it leads Kimi K3 but stays well behind Fable 5 (181 ExploitGym solves) and Sol (216).
Z.ai says the model has been run against real code with Chinese security teams, reporting 2,436 vulnerabilities across 269 open-source projects, 1,097 of them rated critical or high, tracked through a public coordinated-disclosure ledger — 53 CVEs assigned at launch, with the remainder under embargo. The headline anecdote is a “potentially serious” architectural flaw the model found in Cursor, disclosed privately while a fix is prepared. Treat that one carefully: it is Z.ai’s own account, no CVE has been published for it, and there is no independent confirmation (VentureBeat).
The consequence is the structural story. Z.ai delayed the open-weight release by about two weeks — targeted for late August 2026 — explicitly to complete safety evaluation and hardening first (Axios). It echoes OpenAI’s approval-only GPT-5.6-Cyber four days earlier — two labs in one week putting frontier security capability behind a process rather than a price list, from opposite ends of the open/closed divide. As of 19 August 2026 the weights have not shipped and no licence has been published. The GLM-5 line has been MIT throughout, and that is the reasonable expectation for 5.3 — but it is an expectation, not a commitment we can cite.
Pricing and access
Z.ai lists GLM-5.3 at $1.40 input / $4.40 output per million tokens, with cached input at $0.26 — identical to GLM-5.2 and GLM-5.1, so the capability jump arrives at no price increase (Z.ai pricing). Against the frontier cluster that is roughly a third of Claude Opus 5’s $5/$25 and well under GPT-5.6 Sol’s $5/$30.
The API model ID is glm-5.3, served OpenAI-compatible from https://api.z.ai/api/coding/paas/v4, with an Anthropic-compatible endpoint at /api/anthropic that lets it drop into Claude-shaped agent tooling. The flat-rate GLM Coding Plan includes GLM-5.3 on every tier, from about $12.60/month (Lite, annual billing) to $117.60/month (Max), and Z.ai rolled it out to existing subscribers at launch.
Two access caveats. Hosted use routes through Chinese infrastructure, and Zhipu is on the US Entity List — material for data residency, compliance and government procurement. And unlike every previous GLM-5 release, self-hosting is not currently an option, so the usual mitigation isn’t available yet.
How GLM-5.3 compares
- vs the frontier cluster — At 60 on the independent index it sits one point behind Grok 4.6 and GPT-5.6 Sol and three behind Claude Opus 5, while beating Grok 4.6 on both of Z.ai’s hardest agentic-coding rows (Terminal-Bench 3.0 28.3 vs 26; DeepSWE 66.9 vs 65.9) and undercutting it on cost per task. What keeps it below them: no SWE-bench figure, no vision, and no weights to self-host yet.
- vs Kimi K3 — The clearest economic win. The two are level on the independent index (both 60 at max effort on AA’s August re-run), but GLM-5.3 is far cheaper per token ($1.40/$4.40 against $3/$15), cheaper per task ($0.68 vs $0.84), more than twice as fast (84.7 vs 38.2 tokens/sec), and does it at roughly a quarter of K3’s parameter count (~750B against 2.8T) — which would also make it far more practical to serve if the weights land. It also clears K3 on Z.ai’s long-horizon rows (Terminal-Bench 3.0 28.3 vs 17.4, AutomationBench 48.2% vs 46.7%). K3’s advantages are native vision and weights you can download today.
- vs GLM-5.2 — Same base model, same price, same 1M context; large gains on long-horizon coding and cyber evals. The trade is that 5.2 is MIT open weights you can run now, and 5.3 is not.
Known limitations
No SWE-bench figure at all — Pro or Verified, vendor or standardized — which is a real gap for a model marketed on coding, and the main reason we don’t place it higher under a coding-weighted methodology. Text-only: no image input, where Sol, Fable 5, Grok 4.6 and Kimi K3 all take it. Weights withheld: the open release is announced for late August 2026 but hasn’t happened, and no licence is published — so the self-hosting escape from the hosted API’s data-residency issues isn’t available yet. China-hosted API and Entity List considerations apply to all current access. Verbose: Artificial Analysis measured 170M output tokens across its eval against a 72M median, so real bills can run above the headline rate. Vendor cross-model comparisons are Z.ai’s own, using competitor numbers pulled from published cards rather than reruns.
Frequently asked questions
What is GLM-5.3?
GLM-5.3 is Zhipu AI (Z.ai)‘s current flagship model, released 14 August 2026. It uses the same ~750B-parameter mixture-of-experts base as GLM-5.2 with substantially scaled-up post-training, keeps a 1-million-token context, and is built for long-horizon coding agents.
Is GLM-5.3 open source?
Not yet. Every previous GLM-5 release shipped as MIT open weights on day one, but Z.ai withheld GLM-5.3’s weights at launch, citing the model’s faster-than-expected vulnerability-discovery capability, and said the open release would follow after roughly two weeks of safety evaluation. As of 19 August 2026 no weights and no licence have been published.
How much does GLM-5.3 cost?
$1.40 per million input tokens and $4.40 per million output, with cached input at $0.26 — the same rates as GLM-5.2. Artificial Analysis measures $0.68 per completed Intelligence Index task, the lowest in the frontier cluster. The GLM Coding Plan starts at about $12.60/month on annual billing.
Is GLM-5.3 better than Kimi K3?
They are level on capability and far apart on cost. Both score 60 on the Artificial Analysis Intelligence Index at max effort, but GLM-5.3 runs at $1.40/$4.40 against K3’s $3/$15, costs less per completed task ($0.68 vs $0.84), returns tokens more than twice as fast (84.7 vs 38.2 per second), and does it with about a quarter of the parameters. Kimi K3 keeps two real advantages: native vision, and open weights you can download today — which GLM-5.3 does not yet have.
How good is GLM-5.3 at coding?
Strong, with a caveat. Z.ai reports Terminal-Bench 3.0 at 28.3 (up from 4.6) and DeepSWE v1.1 at 66.9 (up from 46.2), which clears Kimi K3 on long-horizon terminal work while trailing Claude Fable 5 and GPT-5.6 Sol on the hardest rows. But Z.ai published no SWE-bench Pro or Verified score in any harness, so the most comparable cross-model coding number simply doesn’t exist for this model.
Why did Z.ai delay the GLM-5.3 weights?
Because the model’s offensive-security capability grew faster than the company planned for during post-training — Z.ai’s own card reports ExploitBench more than doubling, from 24.4% to 54.4% (Z.ai docs) — and the company wanted to complete safety evaluation and hardening before releasing weights that cannot be recalled. OpenAI had gated its GPT-5.6-Cyber model behind an approval programme four days earlier, for the same underlying reason.
Last verified 19 August 2026. The 14 August release date, 1M context, $1.40/$4.40 pricing, model ID and the always-on reasoning change are confirmed against Z.ai’s own documentation and pricing pages; the Intelligence Index score of 60, cost per task and output speed are Artificial Analysis’s independent measurements. The cross-model benchmark table is Z.ai’s launch table, using competitor figures it pulled from published cards rather than reruns — treat it as a ceiling. The Cursor vulnerability is Z.ai’s account with no published CVE and no independent confirmation. The open-weight release is announced for late August 2026 and had not happened at the time of writing; we will update this page when weights and a licence are actually published.