--- title: "Claude Code + Claude Fable 5.2 — run this agent and model — AgentSky" description: "Run Claude Code with Claude Fable 5.2 on AgentSky. See compatibility, current model specifications, pricing boundaries, setup, evaluation steps, and launch the exact pair." url: https://agentsky.dev/agents/claude-code/with/claude-fable-5-2 --- [Log in](https://agentsky.dev/auth/start?returnTo=%2Fagents%2Fclaude-code%2Fwith%2Fclaude-fable-5-2) [Sign up](https://agentsky.dev/auth/start?returnTo=%2Fagents%2Fclaude-code%2Fwith%2Fclaude-fable-5-2&mode=signup) # Claude Code + Claude Fable 5.2 Claude Fable 5.2 is currently selectable with Claude Code on AgentSky. Claude Code controls the work loop and tools; the model supplies reasoning and generation. Start a real task below with both choices already selected. [Repo](https://agentsky.dev/auth/start?returnTo=%2Fagents%2Fclaude-code%2Fwith%2Fclaude-fable-5-2%3Fresume%3D1) How the combination works ## Claude Code runs the work. Claude Fable 5.2 powers each turn. Claude Code runs the coding loop, reads and edits repository files, invokes tools, and reports the result from its cloud workspace. AgentSky supplies the persistent cloud computer, task state, usage records, and enabled tools or connectors around that combination. - Claude Code Owns the agent loop, repository or computer operations, tool calls, and how results are returned. - Claude Fable 5.2 Anthropic's model supplies the reasoning and generated output within Claude Code's supported interface. - AgentSky Keeps the selected pair, workspace, task state, access configuration, and itemized usage together. Current Claude Fable 5.2 facts ### What the model is Claude Fable 5.2 is Anthropic's most capable model for the most demanding technical work, with gains over Fable 5.1 in coding, reasoning, and vision. ### Context window 1M tokens in the current AgentSky catalog. ### Accepted input Text, Image, File in the current AgentSky catalog. ### Model token price $8.80 / $44 per 1M input / output tokens on the displayed AgentSky plan. Runtime and paid tool usage are separate. Evaluate it on real work ## Start with one bounded task and evidence you can inspect. Connect an authorized repository, name the failing behavior or desired change, and define the checks that must pass. Use a disposable branch for an evaluation. 01 ### Open the preselected Composer The Composer above starts with Claude Code and Claude Fable 5.2. Add the repository or project and only the tools the task needs. 02 ### Adapt this first task In this repository, reproduce the reported problem, explain its cause, make the smallest relevant change, and run the affected checks. Summarize the diff and remaining uncertainty. 03 ### Verify the outcome Inspect the actual diff, reproduce the important checks, and look for unrelated edits. Model selection alone does not establish correctness. Fit and limits ## When to choose Claude Code + Claude Fable 5.2 Repository work where Claude Code's coding workflow and supported Claude models fit the team's existing review process. This page establishes - The pair is selectable in the current AgentSky compatibility matrix - The launch carries both selections into the shared Composer - Current catalog facts are shown when the gateway publishes them You still need to measure - Quality on your own repository or workflow - End-to-end task time, recovery effort, and human review - Total task cost, which also depends on runtime, tools, connectors, and token usage ## Claude Code + Claude Fable 5.2 questions ## Related resources [Claude CodeCompare every currently supported model for Claude Code.](https://agentsky.dev/agents/claude-code) [Claude Fable 5.2 APIReview the current model specification, token pricing, and API examples.](https://agentsky.dev/models/anthropic/claude-fable-5-2) [PricingUnderstand model, runtime, tool, and connector charges.](https://agentsky.dev/pricing) ## Run a real task with Claude Code + Claude Fable 5.2. [Run Claude Code + Claude Fable 5.2](https://agentsky.dev/playground/enter?agentType=claude_code&llm=claude-fable-5-2&source=pair-guide) ## Frequently asked questions ### Can Claude Code run Claude Fable 5.2 on AgentSky? Yes. This page is generated only when Claude Fable 5.2 is selectable for new Claude Code runs in AgentSky's current compatibility matrix. If the pair is retired or disabled, the page is removed from discovery and returns not found. ### What does Claude Code do in this combination? Claude Code controls how the agent reads context, invokes tools, works in the cloud computer, and returns progress and results. That behavior remains distinct from the selected model. ### What does Claude Fable 5.2 do in this combination? Claude Fable 5.2 supplies reasoning and generated output for the turns that Claude Code sends through its supported model interface. The current AgentSky catalog lists a 1M-token context window. ### How do I start Claude Code + Claude Fable 5.2? Use the Composer on this page. It already selects Claude Code and Claude Fable 5.2; add a concrete task and any needed project, tools, or connectors. If you are signed out, AgentSky returns you to the same draft after sign-in. ### How much does Claude Code + Claude Fable 5.2 cost? Claude Fable 5.2 is currently displayed at $8.80 / $44 per 1M input / output tokens. That is the model-token component, not a guaranteed task total. Agent runtime, paid tools, connectors, cache usage, and the task's actual token volume can add cost; AgentSky records charges against the same balance. ### What happens if I am signed out or do not have enough balance? The shared AgentSky Composer preserves the task and exact pair through sign-in. Before dispatch, the product checks the account and balance; if funding is required, it shows the billing path instead of silently changing the agent or model. ### Can I switch models without changing Claude Code? Yes, when another model appears in Claude Code's supported model list. Treat the switch as a new evaluation because quality, latency, context handling, tool behavior, and token price may change. ### Is this the same as calling Claude Fable 5.2 through the API gateway? No. A direct gateway call gives you the model response and usage record. This combination also runs Claude Code's tool-using workflow inside an AgentSky cloud computer, so runtime and task behavior are part of the result. ### Is Claude Code + Claude Fable 5.2 the best combination? This page confirms availability and explains how to test the pair; it is not a universal ranking. Compare representative tasks, completion quality, failure recovery, elapsed time, and total recorded cost before choosing a default. --- Markdown copy of https://agentsky.dev/agents/claude-code/with/claude-fable-5-2. Every agentsky.dev page has one: add `.md` to its address or send `Accept: text/markdown`. Start from https://agentsky.dev/llms.txt.