Claude Fable 5.1: what Anthropic's new model can do and what to use in Yenisei
Anthropic released Fable 5.1 on September 1: a million tokens of context, image input, tool calling. What the model can do and which open models in Yenisei suit the same tasks.
Yenisei team
Claude Fable 5.1: what Anthropic's new model can do and what to use in Yenisei
Anthropic released Claude Fable 5.1 on September 1, 2026. Below is what this model is. It is not in the Yenisei catalog: we work with open-weight models and our own Yenisei lineup, and at the end of the article you will find what to use from us for the same tasks.
What this model is
Fable 5.1 is the top model of the Claude lineup, designed for complex development and agent workflows, where a task takes dozens of steps and a mistake at any one of them spoils the result.
- 1M-token context. A whole codebase or a lengthy specification fits in a single request.
- Image input. You can attach a screenshot of an error, an architecture diagram or a mockup to a request.
- Reasoning. Before answering, the model works through the task step by step.
- Tool calling and precise instruction following. This is why the Claude lineup is chosen for autonomous agents.
What to use in Yenisei
- Kimi K2.7 Code, a model for code and development agents: reasoning, tools, image input.
- GLM 5.3, long agent chains and the largest context in the catalog, over 1.3M tokens.
- Qwen3 Coder Next, an inexpensive coding assistant with tool calling.
- Yenisei Code, our coding model with a million-token context.
How to call it
from openai import OpenAIclient = OpenAI( base_url="https://api.yenisei.ru/v1", api_key=YENISEI_API_KEY,)response = client.chat.completions.create( model="kimi-k2.7-code", messages=[{"role": "user", "content": "Analyze this stack trace and suggest a fix"}],)The API is OpenAI-compatible: you change the base URL and the key, and the code stays the same. Payment is in rubles and only for the tokens you use; current prices are on the plans and pricing page.
Who needs this
Those who write code with a model and hit the ceiling: long refactorings, making sense of someone else's project, an agent that runs the tests itself and fixes things based on the results. For everyday questions and texts, a top-tier model is overkill: start with Yenisei 100 or Yenisei Max.