Short answer
- Start with Claude Opus 5.5 for most work. Anthropic's models overview says so, and it costs half as much as Fable 5.2.
- Move to Claude Fable 5.2 for demanding reasoning and long-horizon agent work, or when Opus 5.5 at higher effort still falls short.
- Fable 5.1 is now a legacy model. Fable 5.2 is cheaper ($8 / $40 vs $10 / $50) with the same 1M context and 128k output.
On TeamoRouter today: Opus 5.5 and Fable 5.1 work with one key. Fable 5.2 isn't on TeamoRouter yet; when it launches you only change the model ID. Status: Claude Fable 5.2 page.
New accounts start with a $0 balance — top up before your first request. Top up from $5. If a request returns 402 (400 on the OpenAI format) with "wallet balance is insufficient", your balance is empty — your base URL and key are fine. Top up and retry.
Get one API key for Claude Opus 5.5 and Fable 5.1
Side by side
Official figures from Anthropic's models overview and release notes, checked November 10, 2026:
| Claude Fable 5.2 | Claude Fable 5.1 | Claude Opus 5.5 | |
|---|---|---|---|
| Model ID | claude-fable-5-2 |
claude-fable-5-1 |
claude-opus-5-5 |
| Status | Current (released Nov 9, 2026) | Legacy, still available | Current |
| Input / output per 1M tokens | $8 / $40 | $10 / $50 | $4 / $20 |
| Cache read per 1M tokens | $0.20 (2.5% of input) | $0.25 (2.5% of input) | $0.20 (5% of input) |
| Context window | 1M tokens | 1M tokens | 1M tokens |
| Max output | 128k tokens | 128k tokens | 128k tokens |
| On TeamoRouter | Not yet | Yes | Yes |
Cache reads matter most for coding agents, which reread a large context on every step. On Fable 5.2 and Opus 5.5 they cost the same per token, so the gap between the two comes from uncached input and output.
What changed in Fable 5.2
- Lower price than Fable 5.1, with the same context and output limits.
- Stricter thinking rules: editing earlier turns invalidates its thinking blocks on every account, and only Fable 5.2 and Mythos 5.2 can read its thinking. Claude Code already works within these rules.
- Fast mode on the Claude API: up to 2.5x faster output at $16 / $80 per million tokens.
Details: Claude Fable 5.2 API guide.
Which one to pick
| You want | Pick |
|---|---|
| Everyday coding, refactors, reviews | Claude Opus 5.5 |
| The hardest reasoning and long agent runs, where a mistake costs more than tokens | Claude Fable 5.2 (Fable 5.1 until it's on TeamoRouter) |
| An existing Fable 5.1 setup that edits conversation history | Stay on Fable 5.1 until you've tested the 5.2 prefix check |
A rough cost check: a Claude Code step that reads 200k input tokens and writes 20k output costs about $1.60 + $0.80 = $2.40 on Fable 5.2 and $0.80 + $0.40 = $1.20 on Opus 5.5 at list price, before cache savings. Current TeamoRouter prices are on the pricing page.
FAQ
Is Claude Fable 5.2 better than Opus 5.5?
Anthropic positions Fable 5.2 for demanding reasoning and long-horizon agent work and Opus 5.5 as the default for most workloads. Test both on your own tasks.
Should I move from Fable 5.1 to Fable 5.2?
At list price it's cheaper with the same limits. If your code edits earlier turns, system or tools, run Anthropic's prefix-check test first.