Claude Fable 5 vs Step 3.7 Flash
Claude Fable 5 (Anthropic) and Step 3.7 Flash (StepFun) compared on benchmarks, pricing, context window, and use-case rankings.
- Step 3.7 Flash is cheaper on input tokens — $0.20 vs $10.00 per 1M.
- Claude Fable 5 has a larger context window (1M vs 262K tokens).
- Claude Fable 5 ranks higher for coding (91th vs 42th percentile).
| Attribute | Claude Fable 5 | Step 3.7 Flash |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | 163 | — |
| Coding | 91 | 42 |
| Math | 91 | — |
| Reasoning & Knowledge | 86 | 24 |
| Agentic & Tools | 92 | 97 |
| Specifications | ||
| Developer | Anthropic | StepFun |
| Family | Claude | — |
| Released | Jun 9, 2026 | May 29, 2026 |
| Parameters | — | — |
| Availability | API access | Open weights (unrestricted) |
| Context window | 1M | 262K |
| Price — $/M input | $10.00 | $0.20 |
| Price — $/M output | $50.00 | $1.15 |
| Inputs | text, image, file | text, image, video |
| Outputs | text | text |
| Benchmarks | ||
| AIME 2024/2025 | 100% | — |
| APEX | 64% | — |
| ARC-AGI | 99% | — |
| ARC-AGI-2 | 89% | — |
| GPQA Diamond | 86% | 81% |
| GSO (code optimization) | 78% | — |
| Humanity's Last Exam | 56% | 21% |
| SciCode | 61% | 44% |
| SimpleBench | 82% | — |
| SimpleQA Verified | 71% | — |
| WebDev Arena | 1628 | — |
| WeirdML | 92% | — |
| τ²-bench | 99% | 99% |
Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.
Frequently asked questions
Which is cheaper, Claude Fable 5 or Step 3.7 Flash?
Step 3.7 Flash is cheaper on input tokens at $0.20 per million, versus $10.00 (representative OpenRouter pricing).
Which has a larger context window, Claude Fable 5 or Step 3.7 Flash?
Claude Fable 5 supports up to 1M tokens, compared with 262K for the other.
Which is better for coding, Claude Fable 5 or Step 3.7 Flash?
Across coding benchmarks like SWE-bench Verified and Terminal-Bench, Claude Fable 5 ranks higher — 91th vs 42th percentile among the models tracked on Model Beat.
Want a different match-up? Open the compare tool to add or swap models.
More comparisons
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.