GLM-4.6 is an advanced large language model developed by Zhipu AI (now Z.ai), featuring a state-of-the-art 355 billion parameter Mixture of Experts (MoE) architecture. It is optimized for a broad range of tasks including complex reasoning, coding, writing, and multi-turn dialogue with an extended context window of 200,000 tokens. GLM-4.6 demonstrates industry-leading performance, especially in programming and agentic tasks, making it a top choice for developers and enterprises seeking efficiency and versatility.
GLM-4.6 has been rigorously evaluated across authoritative benchmarks demonstrating competitive or superior results against leading models:
Vs. GLM-4.5: GLM-4.6 offers noticeable improvements in code generation accuracy and maintains a consistent edge in handling ultra-long context inputs, while retaining strong agentic task performance close to GLM-4.5.
Vs. OpenAI GPT-4.5: GLM-4.6 narrows the gap in reasoning and multi-step task accuracy, leveraging its much larger context window; however, GPT-4.5 still leads in raw task precision on some standardized benchmarks.
Vs. Claude 4 Sonnet: While Claude 4 Sonnet excels in coding and multi-agent efficiency, GLM-4.6 matches or surpasses it in agentic reasoning and long-document comprehension, making it stronger for extended-context applications.
Vs. Gemini 2.5 Pro: GLM-4.6 balances advanced reasoning and coding capabilities with enhanced long-form document understanding, whereas Gemini 2.5 Pro is more focused on optimizing individual coding and reasoning benchmarks.
Accessible via AI/ML API. Documentation: available here.
Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "zhipu/glm-4.6", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "zhipu/glm-4.6", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"zhipu/glm-4.6","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| LiveCodeBench | 82.8% | Contamination-free competitive programming problems | Source | July 8, 2026 |
| AIME 2025 | 93.9% | Competition mathematics (AIME), 2025 | Source | July 8, 2026 |
| SWE-bench Verified | 68% | Resolving verified real GitHub issues | Source | July 8, 2026 |
| Intelligence | 18.5 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Coding | 45.8 | Composite score across standardised coding evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Math | 86 | Composite score across standardised mathematics evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
GLM-4.6 This page | Coding + agents | |||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
GLM-4.6 has a 200,000 tokens context window and can return up to 128,000 tokens.
GLM-4.6 takes text as input and returns text.
Use zhipu/glm-4.6 as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
GLM-4.6 became available on October 2, 2025.
GLM-4.6 is priced at input $0.78 / 1M tokens, output $2.86 / 1M tokens, cached input $0.143 / 1M tokens.
Yes, GLM-4.6 can stream responses as they are generated.
GLM-4.6 was built by Zhipu AI.
