Moving to Claude Fable 5.2 on TopxAI
· anthropic, models, api, topxai
Update the Fable model ID, pay 20% less on every price, and check history, computer-use and advisor settings when moving your client to Claude Fable 5.2.
Use claude-fable-5-2 for Claude Fable 5.2 on TopxAI's Claude shared and official routes. Anthropic released it on November 9, 2026, and the upstream ID is the same. It replaces Fable 5.1 outright: the old claude-fable-5-1 ID is retired, returns HTTP 403 and does not forward. Every price is 20% below Fable 5.1.
Update the client and key
Replace the saved model ID in your SDK or coding client. Keep the same base URL and key if the key has no model restriction. For a restricted key, edit its allowed-model list separately; changing the client's model does not expand the key's permissions.
For Claude Code, update ANTHROPIC_MODEL in your shell or ~/.claude/settings.json:
export ANTHROPIC_BASE_URL=https://ai.topxea.com
export ANTHROPIC_AUTH_TOKEN=$TOPXAI_API_KEY
export ANTHROPIC_MODEL=claude-fable-5-2
claude
The Claude Code guide covers persistent settings. For a direct Messages request:
curl https://ai.topxea.com/v1/messages \
-H "x-api-key: $TOPXAI_API_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "Content-Type: application/json" \
-d '{"model":"claude-fable-5-2","max_tokens":4096,"output_config":{"effort":"high"},"messages":[{"role":"user","content":"Review the trade-offs in this API design."}]}'
Prices on both routes
Default USD prices per million tokens, checked against Anthropic's pricing page on November 10, 2026:
| Category | Shared | Official | Anthropic list |
|---|---|---|---|
| Input | $4 | $7.20 | $8 |
| Output | $20 | $36 | $40 |
| Cache read | $0.10 | $0.18 | $0.20 |
| Cache write, 5 minutes | $5 | $9 | $10 |
| Cache write, 1 hour | $8 | $14.40 | $16 |
The shared route is 50% of list and the official route is 90%, including both cache-write durations. There is no long-context premium. Fable 5.2 supports a 1M-token context and up to 128K output tokens. Actual charges use the upstream-reported token categories and the request's captured rates; a cache miss is not automatically a cache write. On the shared route, a Fable 5.2 cache write requested at 5 minutes has so far come back reported as a 1-hour write, and it is charged at the 1-hour price. The model page and usage log show the applicable prices and counts.
Check thinking settings
Adaptive thinking is always on and cannot be turned off. The default effort is high, and Fable 5.2 accepts low, medium, high, xhigh and max. Thinking tokens count toward the output limit and are billed as output, so a very small max_tokens can end with no text.
On Chat Completions and Responses, TopxAI keeps a request that asks to turn thinking off adaptive at the default effort, and translates a manual thinking budget into an effort level. Send a lower effort if you want less reasoning. Native Messages requests are forwarded as sent, and Anthropic documents a 400 for thinking: {"type":"disabled"} and for budget_tokens, so remove both there and set output_config.effort instead.
Read returned content blocks by type, and send native thinking blocks and signatures back unchanged in tool conversations.
Check sampling and tools
Fable 5.2 does not support temperature, top_p or top_k; TopxAI removes them before forwarding on Messages, Chat Completions and Responses. Forced tool selection (any / a named tool, or OpenAI required / a named function) returns 400 without a charge. Use auto.
What can break when you move from Fable 5.1
Anthropic's migration guide lists four changes, and TopxAI forwards all of them unchanged:
- Edited history. If a conversation replays thinking blocks after an earlier turn was changed, Anthropic returns 400. Fable 5.1 enforced this only on newer accounts; Fable 5.2 enforces it on every account. Keep earlier turns byte-for-byte and only append.
- Thinking blocks. Fable 5.1 cannot read blocks written by Fable 5.2.
- Computer use.
computer_20251124returns 400. Usecomputer_toolset_20260801. - Advisor tool. A Fable 5.2 executor accepts only Fable 5.2 or Mythos 5.2 as its advisor.
Ordinary custom tools keep their definitions.
Models: claude-fable-5-2 · Models & pricing
Related
- Claude Fable 5.2 API pricing
- Claude Code with TopxAI: ANTHROPIC_BASE_URL and the key
- Connect your SDK
- How requests are billed
- Moving to Claude Sonnet 5.5 on TopxAI
- Abliterated models on TopxAI: the November 10 prices and how to call them
- GPT-6.1 Terra on TopxAI
Other languages: 简体中文