Zhipu AIGLM 5.3 Prime
A serving tier of GLM 5.3: the same weights on faster inference, at about twice the token price. Benchmark publishers score the base model, so the quality figures to use are its own (index 58.8). What differs here is price and measured speed.
No cheaper ranked model scores higher: it is on the price frontier.
As of 13 Nov 2026, GLM 5.3 Prime by Zhipu AI is not yet ranked on the BenchLeader Index, which needs at least five independent results across two categories. API pricing: $2.80 per 1M input tokens and $8.80 per 1M output tokens ($4.30 blended at 3:1), list price on 13 Nov 2026. It streams about 89 tokens/s (OpenRouter).
Key facts (as of 13 Nov 2026)
- GLM 5.3 Prime API pricing: $2.80 per 1M input tokens and $8.80 per 1M output tokens ($4.30 blended at 3:1) (list price, 13 Nov 2026).
- GLM 5.3 Prime streams about 89 tokens/s (OpenRouter, 13 Nov 2026).
In detail. GLM 5.3 Prime is a Zhipu AI reasoning model. It is a serving tier rather than a separate model — the same weights on faster inference, at about twice the token price — so benchmark publishers score the base model instead and this page carries only its price and measured speed. At $4.30 per million tokens blended it is among the most expensive ranked models. Output speed of 89 tokens per second puts it faster than most, with a first token in 0.7 s.
- Blended price
- $4.30/M
- $2.80 in · $8.80 out
- via OpenRouter
- Output speed
- 89 tok/s
- OpenRouter traffic, 7-day median; not yet measured by Artificial Analysis
- First answer
- 0.70 s
- Context
- 1M
- via OpenRouter
- Full answer
- –
- Cost per task
- –
- Tokens per task
- –
- Released
- –
- date not published by our sources
No independent quality benchmark results yet — only pricing and metadata.
API pricing: where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Alibaba Cloud Int. | 37 tok/s | 0.60 s | $2.80 | $8.80 | 1M | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0038 | – | 4.1 s |
| Summarize a 30-page report | 12,000 / 600 | $0.039 | – | 7.4 s |
| Code edit | 6,000 / 1,500 | $0.030 | – | 17.6 s |
| Agentic coding session | 60,000 / 4,000 | $0.203 | – | 45.6 s |
| Structured extraction | 2,000 / 200 | $0.0074 | – | 3.0 s |
See also
Questions people ask
- Why does GLM 5.3 Prime have no BenchLeader Index score?
- The index needs at least five independent results across at least two categories, and as of 13 Nov 2026 GLM 5.3 Prime does not have them yet. Launch-post scores are never used, so it is ranked once independent publishers have measured it.
- How much does GLM 5.3 Prime cost per 1M tokens?
- As of 13 Nov 2026, GLM 5.3 Prime costs $2.80 per 1M input tokens and $8.80 per 1M output tokens, $4.30 blended (three input tokens per output token).
Data as of 13 Nov 2026. Compare with another model.
Cite as: BenchLeader, “GLM 5.3 Prime: benchmarks, pricing, speed and rank”, data as of 13 Nov 2026, https://www.benchleader.com/models/glm-5-3-prime.