Model tracker / Z.ai
GLM-5.3
GLM-5.3 is an everyday open-weight model from Z.ai, released 18 August 2026. At Z.ai it costs $1.40 per million input tokens and $4.40 per million output tokens, with a 1M-token context window. 10 providers serve it; the cheapest is Baseten at $1.40 input and $4.40 output.
- Made by
- Z.ai
- Released
- 2026-08-18
- Status
- Generally available
- Tier
- Everyday
- Weights
- Open
- Context window
- 1,000,000 tokens
- Max output
- 128,000 tokens
- Input
- text
- API id
glm-5.3
GLM-5.3 price at 10 providers
US dollars per million tokens. Blended is 3 input tokens for every output token, one number to compare on. The cheapest blended price is highlighted.
| Provider | Input | Output | Cached input | Blended | Batch | Checked |
|---|---|---|---|---|---|---|
| Baseten · zai-org/GLM-5.3 | $1.40 | $4.40 | $0.14 | $2.15 | – | 2026-10-14 |
| Cloudflare · @cf/zai-org/glm-5.3 | $1.40 | $4.40 | $0.26 | $2.15 | – | 2026-10-14 |
| Fireworks AI · accounts/fireworks/models/glm-5p3 | $1.40 | $4.40 | $0.26 | $2.15 | 50% off | 2026-10-14 |
| Mistral AI · zai-glm-5-3 | $1.40 | $4.40 | $0.14 | $2.15 | 50% off | 2026-10-14 |
| Modal Labs · zai-org/GLM-5.3 | $1.40 | $4.40 | $0.26 | $2.15 | – | 2026-10-14 |
| Nebius Group N.V. · zai-org/GLM-5.3 | $1.40 | $4.40 | – | $2.15 | – | 2026-10-14 |
| Novita AI · zai-org/glm-5.3 | $1.40 | $4.40 | $0.26 | $2.15 | – | 2026-10-14 |
| Together AI · zai-org/GLM-5.3 | $1.40 | $4.40 | $0.26 | $2.15 | – | 2026-10-14 |
| Z.ai · maker | $1.40 | $4.40 | $0.26 | $2.15 | – | 2026-10-14 |
| AWS · global.zai.glm-5.3 | $1.68 | $5.28 | $0.312 | $2.58 | 50% off | 2026-10-14 |
- Baseten: GLM-5.3 Fast: $2.10 / $6.60.
- Cloudflare: Needs Workers Paid or AI Gateway credits. Billed in Neurons at $0.011 per 1,000; 10,000 Neurons a day free.
- Fireworks AI: US-hosted and Fast variants: $2.10 / $6.60. Priority tier: $1.75 / $5.50.
- Mistral AI: Third-party model hosted by Mistral (since 2026-09-15), served without Mistral modifications. Not yet on Mistral's US regional endpoint.
- Modal Labs: Modal Shared Endpoint, billed per token from us-west; plan credits can't be used for Shared Endpoint usage.
- Nebius Group N.V.: No cached-input or batch price published.
- Z.ai: Cached input storage free for a limited time.
- AWS: Amazon Bedrock, Global cross-region inference, Standard tier. US cross-region: $1.848 in / $0.3432 cached / $5.808 out. 30-minute cache write $2.10. Priority tier +75%; Flex and Batch 50% off.
GLM-5.3 benchmarks
| Benchmark | Score | What it measures | Reported by |
|---|---|---|---|
| LMArena text score | 1478 | Elo-style score from people voting between answers | Independent |
Scores reported by a model's maker aren't independently checked.
Models like GLM-5.3
Other everyday models closest in price, at each maker's own price or the cheapest provider.
| Model | Input | Output | Context | Weights |
|---|---|---|---|---|
| GLM-5.3 | $1.40 | $4.40 | 1M | Open |
| Muse Spark 1.3 Meta | $1.25 | $4.25 | 1M | Closed |
| DeepSeek V4 Pro DeepSeek | $1.32 | $3.96 | 1M | Open |
| Qwen3.8 Max Alibaba | $2.00 | $6.00 | 1M | Open |
| Mistral Medium 3.5 Mistral AI | $1.50 | $7.50 | 256K | Open |
| Grok 4.7 xAI | $2.00 | $6.00 | 500K | Closed |
Compare them all in the model tracker →
GLM-5.3 on dbggr
- Meta's Muse Wonder costs as much as Claude Opus 5.5: $4 in, $20 out per million tokens Daily, 2026-10-12
- OpenAI publishes 722 AI-written math papers, claiming proofs of the quasi-Riemann hypothesis and Unique Games Daily, 2026-10-06
GLM-5.3 FAQ
How much does GLM-5.3 cost?
GLM-5.3 costs $1.40 per million input tokens and $4.40 per million output tokens at Z.ai, and $0.26 per million cached input tokens. Prices are list prices in US dollars, last checked 14 October 2026.
What is GLM-5.3's context window?
GLM-5.3 takes up to 1,000,000 tokens in one request and writes up to 128,000 tokens in one response.
Is GLM-5.3 open weight?
Yes. Z.ai publishes the weights, so you can run GLM-5.3 yourself or at any host that serves it.
Where can I use GLM-5.3?
GLM-5.3 is available from Baseten, Cloudflare, Fireworks AI, Mistral AI, Modal Labs, Nebius Group N.V., Novita AI, Together AI, Z.ai and AWS. At Z.ai, call it as glm-5.3.