Z.ai: GLM 5.3 Prime

GLM-5.3-Prime is the high-speed variant of Z.ai's GLM-5.3, inheriting its full capabilities while delivering 1.5–2× the output throughput through inference acceleration. It supports text input and ou…

z-ai/glm-5.3-prime-20260921

GLM-5.3-Prime is the high-speed variant of Z.ai's GLM-5.3, inheriting its full capabilities while delivering 1.5–2× the output throughput through inference acceleration. It supports text input and output with a 1M-token context window and up to 128K output tokens, and targets coding and agentic workloads, including long-horizon multi-turn agent orchestration, real-time conversation, and streaming code generation.

Reasoning is always on and cannot be disabled. Reasoning efforts `low`, `high`, and `max` are supported; `max` is the default.

Specifications & capabilities

Input
text
Output
text
Context window · tokens
1000000
Capabilities
promptCaching, reasoning, structuredOutputs, toolCalling
Description source ↗

Published prices

1–3 / 3

z-ai/glm-5.3-prime ↗

$0.56

Inference provider
openRouter
Platform
OpenRouter
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions

z-ai/glm-5.3-prime ↗

$2.80

Inference provider
openRouter
Platform
OpenRouter
Unit
input_tokens_per_million
Currency
USD
Billing conditions

z-ai/glm-5.3-prime ↗

$8.80

Inference provider
openRouter
Platform
OpenRouter
Unit
output_tokens_per_million
Currency
USD
Billing conditions

Source specification observations

1–2 / 2

openRouter ↗

field
context_length
value
1000000
normalized value
1000000
scope
source-model
source id
openRouter
confidence
1

openRouter ↗

field
max_output_tokens
value
131072
normalized value
131072
scope
source-model
source id
openRouter
confidence
1

Original sources

1–2 / 2

OpenRouter · Model page ↗

OpenRouter · Catalog ↗

Source links

1–1 / 1

Explore more models

Models

Loading
EVIDENCE / COMPARISON

Benchmarks

Explore every benchmark

Loading

These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

Install TensorTrace