DeepSeek V4.1 Flash 1017 is live, the same day it dropped. Try it now
Sign Up
z-ai

glm-5.2

Open-weight frontier model with strong coding, agentic performance and a 1M context window.

z-ai 1M context
Get API Key View Pricing

Value rank

#10 of 14

intelligence per dollar in our catalogue

Context rank

#1 tied

1M token window

Agentic strength

78%

Terminal-Bench v2.1 task completion

Benchmarks

Independent scores by Artificial Analysis, compared with the strongest models in our catalogue.

Intelligence index

qwen3.8-2.4t-a95b
45.4
deepseek-v4.1-flash-1017
45.1
glm-5.3
44.8
kimi-k3
43.6
glm-5.2
33.7

Coding index

qwen3.8-2.4t-a95b
76.2
kimi-k3
76.2
glm-5.3
74.8
glm-5.2
68.8

Agentic (Terminal-Bench v2.1)

qwen3.8-2.4t-a95b
88.8%
kimi-k3
85%
glm-5.3
83.9%
glm-5.2
77.9%

Long context

kimi-k3
88.7%
deepseek-v4.1-flash-1017
82.7%
qwen3.8-2.4t-a95b
80.3%
glm-5.3
79.7%
glm-5.2
78.3%

Best for

Where this model earns its keep.

Agentic pipelines and tool use Long-document analysis Coding and engineering Prompt-cached workloads

The numbers

Pricing is live from our platform. Prices per 1M tokens, zero data retention on every request.

Input price$1.50
Cache read price$0.38
Output price$4.50
Context window1M tokens
Intelligence / coding index33.7 / 68.8
Agentic: Terminal-Bench v2.1 / tau278% / 99%
Long-context reasoning78%
GPQA / MMLU-Pro90% / -

Or consider

Close alternatives in the catalogue.

qwen3.8-27b

Intelligence 33.7 · agentic 80%

deepseek-v4-flash-0731

Intelligence 34.3 · agentic 79%

deepseek-v4-pro

Intelligence 36 · agentic 79%

Quick start

OpenAI-compatible. Switch in one line.

# pip install openai
client = OpenAI(base_url="https://api.tensorx.ai/v1", api_key="tsx-...")
r = client.chat.completions.create(
    model="z-ai/glm-5.2",
    messages=[{"role": "user", "content": "Hello"}],
)

Benchmark data from the Artificial Analysis Intelligence Index, measured independently. Pricing live from the TensorX platform. All inference on EU-sovereign infrastructure with zero data retention.