GLM-5.3 — Specifications
Compiled by Mustafa Ihsan from the vendor’s published documentation · Last updated
| Developer | Zhipu AI |
|---|---|
| Type | LLM (coding/agentic, MoE) |
| Modality | Text → Text |
| Parameters | 744B total / 40B active (MoE) |
| Context window | 1M |
| Max output | 128K |
| License | GLM-5.3 License (open weights) |
| Open weights | Yes |
| Released | 2026-08 |
| Input price | $1.40 /1M |
| Output price | $4.40 /1M |
| API providers | Zhipu (Z.ai), Alibaba Cloud Model Studio, OpenRouter |
Benchmarks
| Z.ai-reported | Terminal Bench 3.0 28.3 (GLM-5.2: 4.6); DeepSWE v1.1 66.9 (GLM-5.2: 46.2); CyberGym 84.5 (GLM-5.2: 77.2). Cached input $0.26 per 1M tokens. |
|---|
What is GLM-5.3?
GLM-5.3 is Zhipu AI’s (Z.ai) latest flagship model, released on 18 August 2026. According to Z.ai, it uses the same base model as GLM-5.2, with every improvement coming from post-training: Z.ai reports a 50% gain over GLM-5.2 on its in-house Z.ai Code Bench and much stronger results at finding and exploiting software vulnerabilities, including 84.5% on CyberGym. It is a 744B-parameter mixture-of-experts with 40B active parameters. It takes text only, with a 1M-token context window and up to 128K output tokens, and it always reasons: effort can be set to low, high or max, but reasoning cannot be switched off.
Z.ai’s API charges $1.40 per million input tokens, $0.26 per million cached input tokens and $4.40 per million output tokens, with cache storage free for a limited time, the same price as GLM-5.2. Alibaba Cloud Model Studio also lists it at $1.40 / $4.40, and it is available to subscribers of Z.ai’s GLM Coding Plan. For cheaper work, GLM-5.3-Flash (320B total, 18B active) costs $0.15 in / $0.50 out.
The weights are on Hugging Face in FP8, a download of about 756 GB, under Z.ai’s GLM-5.3 License rather than the MIT licence of GLM 5.2. The licence is permissive, but Model-as-a-Service companies with more than $10 billion in revenue over 12 months must pass a Z.ai security review before commercial use. At the same price, GLM-5.3 is the natural pick over GLM 5.2 for new projects; compare monthly costs with the AI API cost calculator.
GLM-5.3 pricing: API cost per 1M tokens
| Input (per 1M tokens) | $1.40 |
|---|---|
| Output (per 1M tokens) | $4.40 |
| Output/input ratio | 3.1× |
| Blended (4:1 in:out) | $2.00 per 1M tokens |
What GLM-5.3 costs per month
Real monthly spend at a 4:1 input-to-output mix — the ratio a typical chat or RAG workload actually produces.
| Workload | Tokens / month | Cost / month |
|---|---|---|
| Side project | 1M in / 0.25M out | $2.50 |
| Small team | 20M in / 5M out | $50 |
| Production | 200M in / 50M out | $500 |
Run your own numbers in the AI API cost calculator.
Cheaper alternatives to GLM-5.3
| Model | Blended $/1M | You save |
|---|---|---|
| Llama 3.1 8B open | $0.0240 | 99% cheaper |
| Mistral NeMo 12B open | $0.0292 | 99% cheaper |
| Gemma 3 4B open | $0.0600 | 97% cheaper |
Frequently asked questions
How much does GLM-5.3 cost per 1M tokens?
GLM-5.3 costs $1.40 per 1M input tokens and $4.40 per 1M output tokens. At a typical 4:1 input-to-output mix that blends to about $2.00 per 1M tokens.
How much does GLM-5.3 cost per month?
A small-team workload of 20M input and 5M output tokens a month costs about $50 on GLM-5.3. A side project (1M in / 0.25M out) costs roughly $2.50.
What is a cheaper alternative to GLM-5.3?
Llama 3.1 8B is the strongest cheaper option in our database at $0.0240 per 1M blended — about 99% less than GLM-5.3. It is also open-weight, so self-hosting is an option.
Can I run GLM-5.3 locally?
Yes. GLM-5.3 is open-weight and needs about a large amount of VRAM at 4-bit quantisation.
Why does GLM-5.3 charge more for output than input?
Output tokens are generated one at a time and cannot be batched the way a prompt can, so they cost the provider more to serve. GLM-5.3 charges 3.1× more for output, which is why prompt-heavy workloads are far cheaper to run than generation-heavy ones.
Prices are the published list rates for the model's primary API and are reviewed as providers change them. Volume, batch and cached-input discounts are not included. Compare every model side by side in the AI models database or the LLM leaderboard.