Zukko Blog
Guides to Claude Code, Cursor, Codex and the Claude and GPT APIs: setup, token savings and troubleshooting.
Claude API: what it is and how it differs from the claude.ai chat
5 min · · API keys
The Claude API gives your code access to the models and is billed per token. The Messages API, caching and tools explained, with a first request.
AI agent: what it is, how it is built and how it differs from a chatbot
5 min · · агенты
An agent is a model that picks its own actions and repeats a loop until the task is done. The structure, examples and honest limits.
AI for 1C: how to write code in the built-in language with a model
5 min · · development
A model cannot see your configuration and may invent methods and attributes. Here is how to give them metadata and examples, verify the code and estimate request costs.
Claude Opus in Russia: how to use Opus 5.5 in chat and via API
5 min · · models
Three working ways to use Claude Opus 5.5 from Russia: web chat, an API request and Claude Code. With code and a cost estimate at October 18, 2026 prices.
Neural network hallucinations: why a model makes things up and how to reduce it
5 min · · development
Made-up facts cannot be removed completely, but they can be made rare and visible. The causes and five measures with working Python code.
Translating Russian to English with a neural network: how to get natural English
5 min · · автоматизация
Good English depends on parameters people often forget to set: audience, register, language variant. A prompt, back-translation checks and the price of a batch.
Translating English to Russian with a neural network: model, prompt and checks
4 min · · автоматизация
Natural Russian comes from an instruction, a glossary and automated checks, not from picking a single "best" model. A scheme with code and a price comparison.
A neural network for programming online: a browser chat with no install
5 min · · development
You do not need an editor to get started: a browser and a key are enough. Working prompts, model choice and the price of a typical conversation.
Ollama: what it is, what it's for and how it differs from a cloud API
6 min · · models
Ollama runs open language models on your computer and serves them over a local API. What it can and can't do, and when a cloud API is the better fit.
The context window in neural networks: what it is and why size is not everything
5 min · · models
The context window is a model's entire working memory for one request: prompt, history, files and the answer. How to count it and what a long context costs.
LM Studio vs Ollama: which to pick for running local models
6 min · · models
Both programs run models on your computer and expose an OpenAI-compatible API. We cover the differences and count when local running loses to the cloud.
Which AI is best for lawyers: tasks, risks and checking answers
6 min · · бизнес
There is no universal best AI for lawyers, but there is a method of choosing. We cover tasks, hallucination and data leak risks, and a way to check answers.
GLM API free: what is really free and what GLM 5.3 Flash costs
6 min · · models
Free offers around GLM exist, but with caveats. We look at what usually hides behind the word free and calculate how many requests one dollar buys on GLM 5.3 Flash.
Claude Haiku vs Sonnet vs Opus: how the lines differ and what the API has
5 min · · models
Haiku, Sonnet and Opus set three levels of size, price and strength. We explain the lines, the Claude list in the gateway API and a light-model replacement when Haiku is not in the catalog.
Fine-tuning or RAG: what to choose to make a model work with your data
6 min · · models
Fine-tuning changes how a model behaves, while RAG and long context give it facts. We compare three approaches by cost, effort and freshness and give a rule for choosing.
RAG explained in plain words and when you really need it
5 min · · models
RAG adds found fragments of your documents to the request to the model. We explain the design without jargon and calculate when long context is simpler.
Qwen API key: how to get one from Alibaba Cloud or via a gateway
5 min · · API keys
Two ways to a Qwen key: the official one in Model Studio and a single gateway key. Steps, addresses, model ids and a sample request.
DeepSeek context window: its size and how to use it without overpaying
5 min · · models
The window size of current DeepSeek models in Zukko, how it differs from the answer length, and how to measure a request in advance and price it without surprises.
GLM API key: how to get a key for GLM 5.3 and which URL to use
5 min · · API keys
A GLM key can come from the Z.ai platform or from a gateway. We show both routes, the correct addresses, model ids and a cost comparison on an example.
GLM-5.3-Flash vs GPT-5.6-Luna: the Cheapest Models in the API
5 min · · сравнение
The two cheapest models on Zukko differ in price, context window, and /v1/responses support. We do the cost math and show how to compare them on your data.
GPT-5.6 Sol or Claude Sonnet 5? Same price on Zukko, how to choose
4 min · · models
On Zukko both models cost the same and work in either API format. What actually differs, and how to compare the models on your own tasks.
Sonnet 5 or Opus 5.5? Same price on Zukko, so test on your own tasks
4 min · · models
On Zukko both models cost the same, so you choose by results on your tasks. Here is how to run an A/B test and what to measure.
Claude Opus 5.5 overview: API access, pricing and testing on your tasks
5 min · · models
How to call Claude Opus 5.5 via Zukko, what a request costs by formula, and how to decide on your own tasks whether you need this model.
Error 503 model_disabled: What It Means and How to Handle It
4 min · · errors
model_disabled means one specific model is switched off right now, not that the whole service is down. How to find a replacement, switch your IDE, and code a fallback.