Claude model for creative writing, analysis, and controlled agent workflowsClaude Fable 5 by anthropic is offered by 62 providers on this page. Public prices are shown for 60 of them. Its official list price is $10.00 per 1M input tokens and $50.00 per 1M output tokens. The lowest paid channel is Xpersona at $3.00 / $18.50 per 1M, about 2.9× below the list price. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it. It also has 1 free ($0) channel; free tiers usually carry rate limits, and subscription-covered access bills $0 per token only after the subscription fee.
Median output speed is 66 tokens per second, with 107.71s to the first token.
The context window is 1,000,000 tokens, with an output limit of 128,000 tokens. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, and PDF. Its training knowledge cuts off at 2026-01-31.
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| GitLab Duo duo-chat-fable-5 | Cloud | Free | $0 | $0 | 1,000,000 | 128,000 | ||
| Xpersona | Gateway | $3.00 | $18.50 | $0.30 | — | 1,000,000 | 128,000 | |
| xxapi | Gateway | $3.85 | $19.23 | $0.38 | $4.81 | 1,000,000 | 128,000 | |
| Outer Labs | Gateway | $6.00 | $30.00 | $0.60 | $7.50 | 1,000,000 | 128,000 | |
| KoRouter | Gateway | $6.50 | $32.50 | $0.65 | $8.13 | 1,000,000 | 128,000 | |
| Kunavo | Gateway | $7.00 | $35.00 | $0.70 | $8.75 | 1,000,000 | 128,000 | |
| AnthropicOfficial | First-party | $10.00 | $50.00 | $1.00 | $12.50 | 1,000,000 | 128,000 | |
| Perplexity Agent | Gateway | $10.00 | $50.00 | $1.00 | $12.50 | 1,000,000 | 128,000 | |
| CrossModel | Gateway | $10.00 | $50.00 | $1.00 | $12.50 | 1,000,000 | 128,000 | |
| LLM Gateway | Gateway | $10.00 | $50.00 | $1.00 | $12.50 | 1,000,000 | 128,000 |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Interleaved thinking (reasoning between tool calls) is declared by 7 of 62 providers.
6 / 62 providers expose no reasoning control (reasoning_options: []).
1 more channels offer $0 (GitLab Duo); free tiers usually have rate limits and no SLA, excluded from ranking.
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| SWE-Bench Pro | — | 80.3 | resolve rate | Source ↗ |
| SWE-Bench Verified | — | 95 | resolved | Source ↗ |
| Terminal-Bench | v2.1 | 88 | success rate | Source ↗ |
| Humanity's Last Exam | variant: no tools | 59 | accuracy | Source ↗ |
| Humanity's Last Exam | variant: with tools | 64.5 | accuracy | Source ↗ |
| OSWorld-Verified | — | 85 | success rate | Source ↗ |
| FrontierCode | variant: high effort · dataset: Diamond | 29.3 | pass rate | Source ↗ |
| GDPval-AA | — | 1932 | Elo | Source ↗ |
| AutomationBench | — | 17.4 | success rate | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.