Selection guide
Which Anthropic model fits the workload?
Choose Claude 3.7 Sonnet when its lower listed prices and documented hybrid-reasoning workflow fit the task. Evaluate Claude Fable 5 when million-token context and long-running coding or knowledge-work agents justify a higher budget.
API + app
Claude 3.7 Sonnet
Consider for: Code maintenance and multi-step document work using controllable extended thinking at lower listed token prices.
Account for: Smaller context and older source snapshot than the Fable record.
API + app
Claude Fable 5
Consider for: Premium long-running coding and knowledge-work agents that can use a million-token context window.
Account for: Substantially higher listed input and output prices.
Anthropic model specifications and prices
| Model | Access | Released | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| Claude 3.7 Sonnet | API + app | February 24, 2025 | 200K | $3.00 | $15.00 |
| Claude Fable 5 | API + app | June 9, 2026 | 1.0M | $10.00 | $50.00 |
Prices are provider-listed USD snapshots per million tokens where represented. Caching, batch processing, long-context rules, modalities, regions, and service tiers can change effective cost.
Reported benchmark coverage
| Model | MMLU-Pro | GPQA Diamond | SWE-bench Verified | LiveCodeBench | SWE-Bench Pro | Artificial Analysis Intelligence Index |
|---|---|---|---|---|---|---|
| Claude 3.7 Sonnet | 84.1 | 84.8 | 70.3 | 46.4 | — | — |
| Claude Fable 5 | — | 92.6 | — | — | 80 | 59.9 |
A dash means the exact evaluation was not retained for that model. Unlike benchmark families and source settings are not merged into an overall score.
Provider and model sources
Each model keeps its own source record and limitations. Review the provider API reference for current authentication, quotas, SDK status, and pricing rules before implementation.