GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering, and complex multi-step automation. Reasoning efforts high and xhigh are supported; xhigh maps to max reasoning.

How much does GLM 5.2 cost?

GLM 5.2 costs $0.50/M input tokens and $3.15/M output tokens, with separate rates for Cache Read at $0.115/M tokens.

What is the context length of GLM 5.2?

GLM 5.2 has a 1,048,576 token context window. It supports up to 131,072 completion tokens.

Does GLM 5.2 support tool calling and structured outputs?

Yes. GLM 5.2 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Which providers serve GLM 5.2?

GLM 5.2 is served by 26 providers on OpenRouter: Sail Research, Ambient, Decart, DigitalOcean, StreamLake, NovitaAI, GMICloud, DeepInfra and 18 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

When was GLM 5.2 released?

GLM 5.2 was released on June 16, 2026.

GLM 5.2 - API Pricing & Benchmarks

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), or Exacto (highest tool-calling accuracy).

Latency / throughput


	$0.50	$3.15	$0.115	1.29s	16 tps	99.55%
	$0.60	$2.00	$0.15	9.02s	6 tps	85.37%
43% off	$1.20$0.684	$4.00$2.28	$0.20$0.114	2.34s	39 tps	99.33%
	$0.70	$2.20	$0.105	0.87s	81 tps	99.11%
48% off	$1.40$0.7336	$4.40$2.306	$0.26$0.1362	1.95s	41 tps	99.85%
47% off	$1.40$0.7406	$4.40$2.328	$0.26$0.1375	2.03s	41 tps	99.94%
47% off	$1.40$0.742	$4.40$2.332	$0.26$0.1378	1.80s	45 tps	99.31%
	$0.75	$2.40	$0.14	1.17s	42 tps	98.93%
	$0.75	$2.90	$0.17	0.49s	95 tps	96.22%
	$0.76	$2.42	$0.14	0.65s	90 tps	99.80%
	$0.966	$3.036	$0.1932	1.13s	49 tps	99.50%
	$1.10	$4.10	$0.22	1.24s	38 tps	98.41%
15% off	$1.40$1.19	$4.40$3.74	$0.26$0.221	2.04s	39 tps	99.96%
	$1.26	$3.00	$0.22	1.82s	33 tps	99.84%
	$1.26	$3.96	$0.234	6.24s	33 tps	97.95%
10% off	$1.40$1.26	$4.40$3.96	$0.26$0.234	2.19s	42 tps	98.05%
	$1.40	$4.40	$0.26	5.23s	26 tps	99.09%
	$1.40	$4.40	$0.14	1.56s	40 tps	98.06%
	$1.40	$4.40	$0.26	3.69s	36 tps	98.87%
	$1.40	$4.40	$0.26	0.64s	109 tps	99.80%
	$1.40	$4.40	$0.26	0.84s	104 tps	99.87%
	$1.40	$4.40	$0.26	2.16s	31 tps	99.06%
	$1.40	$4.40	$0.26	1.45s	54 tps	99.88%
	$1.40	$4.40	$0.26	0.83s	54 tps	99.83%
	$1.40	$4.40	$0.14	1.88s	56 tps	99.73%
	$2.10	$6.60	$0.21	0.80s	104 tps	98.92%
	$2.10	$6.60	$0.21	1.77s	50 tps	99.86%
	$2.10	$6.60	$0.21	1.88s	66 tps	99.74%
	$2.31	$7.26	$0.462	1.48s	38 tps	99.91%
	$1.40	$4.40	$0.26	0.67s	75 tps	94.22%

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

109tok/s

P50, best across providers

Latency

0.49s

P50, best provider

Latency / throughput

AutoExacto Benchmarks

GPQA DiamondTAU-Bench

Wafer Fast

92.2%73.3%

Venice

84.8%77.9%

Baidu Qianfan

85.9%76.8%

Z.ai

86.0%75.9%

Ambient

84.0%77.8%

Wafer

--75.3%

+26 more providers

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)

100.00%

Availability (3d)

99.80%

Availability over the last 3 days

Last 72 hours

Availability 99.80%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.70%

Without Routing

88.86%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and customization options.

Benchmarks

Scores on standardized evaluations. Higher percentages are better — and rank percentile shows where this model lands among all models on OpenRouter.

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

1.

Claude Code

Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.

4.15Ttokens

2.

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

1.59Ttokens

3.

schema-markup-generation

new

774Btokens

4.

pi

There are many coding agents, but this one is yours.

263Btokens

5.

OpenClaw

OpenClaw is an open-source AI agent that connects to your messaging apps and takes real actions on your behalf, from running commands and browsing the web to managing files and sending emails.

239Btokens

Activity

Token volume and request traffic to this model over time.

Quick Start

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Z.ai: GLM 5.2

z-ai/glm-5.2

Z.ai: GLM 5.2

z-ai/glm-5.2

Providers

Pricing

Performance

Uptime

Benchmarks

Apps

Activity

Quick Start

About Z.ai: GLM 5.2

More models from Z.ai

Frequently asked questions

Providers

Pricing

Performance