GLM 5.3 API — unlimited & flat-rate access

GLM 5.3 (also written “GLM-5.3”) is Z.ai’s (Zhipu AI) frontier coding and agentic model — the same 753B-parameter base as GLM 5.2 with scaled-up post-training, released on August 14, 2026 and open-weighted on August 28. CheapestInference serves it in the Frontier Pool through an OpenAI- and Anthropic-compatible API on unlimited time-block subscriptions — so your cost does not scale with tokens. It replaced GLM 5.2 in place on August 30, 2026; existing Frontier subscriptions got the upgrade with nothing to change.

Quick facts


Model	GLM 5.3
Model ID	`glm-5.3`
Version	Served since 2026-08-30 — in-place upgrade of the Frontier Pool’s GLM slot (previously GLM 5.2)
Context window	1M tokens
Max request size	8 MB per request
Reasoning	Reasoning model: effort levels `low` / `high` / `max` — see Reasoning
Open weights	Yes — zai-org/GLM-5.3 on Hugging Face (fp8, 141 shards), published August 28, 2026
License	GLM-5.3 License (custom, not MIT like GLM 5.2): commercial use allowed with attribution; operators of a Model-as-a-Service business with more than US$10B aggregate revenue over any 12 months must pass Z.ai’s security review first — license text
Per-token price elsewhere	$1.40 in / $4.40 out per 1M (cached input $0.26) — reference only, unchanged from GLM 5.2
Endpoints	`/v1/chat/completions` (OpenAI), `/anthropic/v1/messages` (Anthropic)
Pricing	From $71/mo — reserve an 8-hour daily time block, up to full 24/7

Call GLM 5.3

from openai import OpenAI

client = OpenAI(
    base_url="https://api.cheapestinference.com/v1",
    api_key="sk-..."  # your subscriber key
)

response = client.chat.completions.create(
    model="glm-5.3",
    messages=[{"role": "user", "content": "Write a unit test for..."}],
)

curl https://api.cheapestinference.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "glm-5.3", "messages": [{"role": "user", "content": "Hello"}]}'

GLM 5.3 is a reasoning model — it produces internal reasoning before its final answer. Set a generous max_tokens so the response is not truncated mid-reasoning.

What changed from GLM 5.2

Same base model, more post-training. On Z.ai’s published numbers the step is largest in exactly the workloads the Frontier Pool is bought for — agentic coding (Terminal-Bench 2.1 81.0 → 88.2, SWE-Marathon 19.4 → 42.5) and tool-driven automation (Toolathlon 59.9 → 73.0) — plus a headline jump in cyber-security capability (CyberGym 84.5). The full analysis, the vendor benchmark table and the head-to-head with GLM 5.2 are in GLM-5.3: specs, benchmarks, pricing & API options.

For subscribers the upgrade was transparent: the Frontier Pool’s GLM slot now serves GLM 5.3 at the same flat price. Requests that still send model: "glm-5.2" are served by GLM 5.3 until September 30, 2026, then return an invalid-model error — switch to glm-5.3 when convenient. See GLM 5.2 — retired.

Why flat-rate GLM 5.3

GLM 5.3 is a capable, frontier coding and reasoning model. On CheapestInference you pay a fixed monthly fee rather than per token, so heavy coding and agent workloads have a predictable cost. It is part of the Frontier Pool alongside MiniMax M3, and works in any OpenAI-compatible client (Claude Code, Cline, Roo Code, Continue, and similar).

Common questions

Is there a GLM 5.3 API? Yes. Use model id glm-5.3 against https://api.cheapestinference.com/v1. The API is OpenAI- and Anthropic-SDK compatible.

How much does GLM 5.3 cost? From $71/month ($60.35/mo billed annually). You reserve one or more 8-hour daily time blocks (up to full 24/7) and use GLM 5.3 with no token caps — billed at a flat monthly fee, not per token.

Who makes GLM 5.3? GLM 5.3 is made by Zhipu AI (Z.ai). CheapestInference serves it on flat-rate, unlimited time-block subscriptions.

Is GLM 5.3 unlimited on CheapestInference? Yes — during your reserved time blocks there are no token caps: use GLM 5.3 as much as you want for the same flat monthly fee.

Is GLM 5.3 open-weights? Yes, since August 28, 2026 — under Z.ai’s custom GLM-5.3 License (commercial use allowed with attribution; a security-review clause applies only to Model-as-a-Service operators above US$10B revenue). GLM 5.2 was MIT; the terms did not carry over.

I still send glm-5.2 — do I need to change anything? Not immediately. glm-5.2 requests are served by GLM 5.3 until September 30, 2026. After that, the id returns an invalid-model error, so update to glm-5.3 before then. GET /v1/models is always the authoritative live list.