GLM-5.3: Benchmarks, Pricing & Context Window
GLM-5.3 is a language model from ZAI, released in August 2026, with a 1.0M-token context window, and pricing from $1.40/M input, $0.260/M cached input, $4.40/M output.
GLM-5.3 is Z.AI's open-weight flagship for complex coding, long-horizon agentic work, and cybersecurity tasks. It uses the same base model as GLM-5.2, with its capability gains coming from post-training. The Hugging Face checkpoint is a
GLM-5.3 benchmarks
Capability tiers
Standing within each category, adjusted for leaderboard depth.
Real tasks performance
High-confidence performance for GLM-5.3 across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.
Performance by conversation depth
How GLM-5.3 holds up as conversations get longer.
Quality Tracker
GLM-5.3 Performance Across Datasets
Scores sourced from the model's scorecard, paper, or official blog posts
llm-stats.com - Wed Sep 09 2026GLM-5.3 pricing
Providers
GLM-5.3 starts at $1.20 per million input tokens and $4.00 per million output tokens via DeepInfra. Reused prompt prefixes cost $0.120 per million cached input tokens. See all 4 providers below with their per-token pricing, latency, throughput, and modality support.
| Provider | Input $/M | Cached input $/M | Output $/M | Context in / out | TTFT p95 s | Output p5 c/s | Modalities in / out |
|---|---|---|---|---|---|---|---|
 DeepInfra | $1.20 | $0.120 | $4.00 | 1.0M/1.0M | — | — | / |
 FriendliAI | $1.40 | $0.260 | $4.40 | 1.0M/131.1K | — | — | / |
 Novita | $1.40 | $0.260 | $4.40 | 1.0M/131.1K | — | — | / |
 ZAI | $1.40 | $0.260 | $4.40 | 1.0M/131.1K | 6.27 | 48 | / |
Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests.
GLM-5.3 model size
GLM-5.3 has 753 billion parameters. See how it compares to other models in the same parameter range.
GLM-5.3 context window
Input and output token limits for GLM-5.3, plus how it ranks on long-context understanding.
Try now
huggleMake it with
GLM-5.3.
GLM-5.3 latency
GLM-5.3 time to first token, sustained output throughput, and failed-request rate from live model usage over the trailing 7 days.
Provider operational metrics
Time to first token, output throughput, and failed-request rate from live model usage
GLM-5.3 examples
Recent arena outputs from GLM-5.3, picked from the highest-ranked matchups.
GLM-5.3 license
GLM-5.3 is released under the GLM-5.3 License license, which permits commercial use, has 753.0B parameters.
- License
- GLM-5.3 License
- Commercial use allowed
- Parameters
- 753.0B
Open-weights license permitting use, modification, distribution, sublicensing, sale, deployment, and fine-tuning. Model-as-a-Service businesses with more than US$10B in aggregate revenue over any consecutive 12-month period must pass Z.AI's security review before commercial use. The license is source-available/open-weight rather than an OSI-approved open-source license.
GLM-5.3 resources
Official sources for GLM-5.3: provider documentation, official playground, paper or system card, official launch post, source repository, model weights.
GLM-5.3 vs other models
The most-compared alternatives to GLM-5.3 are Claude Mythos Preview, Qwen3.8 Max, Muse Spark 1.1. Open any pair side-by-side for benchmarks, pricing, context, and latency.
Models like GLM-5.3
Models ranked just above and below GLM-5.3 by LLM Stats score.
FAQ
Common questions about GLM-5.3.