GLM 4.7 Flash
Neural Network
GLM 4.7 Flash for code and agentic scenarios: a 30B class Z.ai model balancing quality and speed.
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 200 000 tokens
- Max answer length
- 117 964 tokens
- Reasoning effort
- low, medium, high
- Capabilities
- reasoning, function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
import OpenAI from 'openai';
const openai = new OpenAI({
apiKey: '<your bothub access token>',
baseURL: 'https://openai.bothub.chat/v1'
});
// Sync - Text generation
async function main() {
const chatCompletion = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'glm-4.7-flash',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'glm-4.7-flash',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works GLM 4.7 Flash?
GLM 4.7 Flash is a 30B class model from the Z.ai GLM family, designed to balance response quality and speed. Compared to the previous version, programming skills and multi-step task planning have been enhanced, making it ideal for agentic scenarios involving chains of actions rather than single responses. In BotHub, you can use it without a VPN or foreign card: payments are made with Russian cards in rubles, charged per token, and unused balances do not expire. Other aggregator models are available in the same window, and a unified OpenAI-compatible API allows you to switch models in your code without rewriting integrations; requests are encrypted with AES-GCM. Typical tasks include writing and refactoring code, debugging errors and test failures, building agents that execute multi-step plans, processing high-volume requests where token efficiency is key, and connecting the model to internal services via API. For companies, BotHub provides contracts, invoices, EDI, an admin panel, and usage limits.Frequently asked questions about GLM 4.7 Flash
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
The model works with text: it answers questions, writes and analyzes code, processes documents and images, reasons through tasks before answering, and calls external functions. A 200,000-token context allows you to keep a large project or a batch of files in a single dialogue.
Up to 117,964 tokens in a single response — about the size of a small book. This is enough for long technical texts, detailed code analysis, or entire documentation without splitting it into parts or asking to continue. The total dialogue window is 200,000 tokens.
Yes, reasoning before answering is confirmed by our data: the model breaks down the task into steps first and only then formulates the result. This helps with math, logic, code debugging, and scenarios where a chain of reasoning is more important than a quick, random reply.
Yes, both. The model calls functions and outputs structured JSON, making it convenient for agentic scenarios and connecting to external services. Through BotHub's unified OpenAI-compatible API, this can be connected without rewriting your integration.
Images and documents — yes, you can upload them and get a text analysis: diagrams, screenshots, tables, contracts. Audio and video are not supported in our data, so it is better to use a specialized model in BotHub, which you can switch to in the same window.
We do not have specific data on the quality of Russian for this model, so we cannot make any promises — please test it with your specific task. In BotHub, this is easy: pay-as-you-go per token without a subscription, and other models are available in the same window for comparison.