# Claude Fable 5 vs Claude Sonnet 4.6

> Side-by-side benchmark comparison for Claude Fable 5 and Claude Sonnet 4.6 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/claude-fable-vs-claude-sonnet-4-6
- Last updated: September 4, 2026

- Shared sourced benchmarks: 10
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Pick Claude Fable 5 if you want the stronger benchmark profile. Claude Sonnet 4.6 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- Claude Fable 5 leads overall 80.9 to 64.21.
- The clearest category separation is in instruction following, where the averages are 78.3 for Claude Fable 5 and 47.9 for Claude Sonnet 4.6.
- The biggest single benchmark swing is FrontierCode 1.1 Main in Coding, with scores of 53.5% and 24.3%.
- Claude Sonnet 4.6 is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- Claude Fable 5 also has the larger context window at 1M+.

## Model Snapshot

| Property | Claude Fable 5 | Claude Sonnet 4.6 |
|----------|----------|----------|
| Creator | Anthropic | Anthropic |
| Type | Proprietary | Proprietary |
| Reasoning | Reasoning | Non-Reasoning |
| Context | 1M+ | 200K |
| Overall Score | 80.9 | 64.21 |
| Benchmarks Covered | 18 | 25 |
| Pricing (input/output) | $10.00 / $50.00 | $3.00 / $15.00 |

## Category Breakdown

### Agentic

- Winner: Claude Fable 5
- Claude Fable 5 public-lane score: 74.8 (Supported · #3/151)
- Claude Sonnet 4.6 public-lane score: 45 (Supported · #96/151)

| Benchmark | Claude Fable 5 | Claude Sonnet 4.6 | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 3.0 | 34.0% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | 84.3% | 59.1% | Claude Fable 5 |
| OSWorld-Verified | 85% | 72.1% | Claude Fable 5 |
| Terminal-Bench 2.1 (Vals) | 80.5% | 57.3% | Claude Fable 5 |
| Claw-Eval | Coming soon | 67.8% | Coming soon |
| CyberGym | Coming soon | 65.2% | Coming soon |
| Gert Labs | Coming soon | 62.92% | Coming soon |
| OSWorld 2.0 | Coming soon | 8.3% | Coming soon |
| JobBench | Coming soon | 36.9% | Coming soon |

### Coding

- Winner: Claude Fable 5
- Claude Fable 5 public-lane score: 76.9 (Supported · #2/183)
- Claude Sonnet 4.6 public-lane score: 52.4 (Supported · #55/183)

| Benchmark | Claude Fable 5 | Claude Sonnet 4.6 | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 95% | 79.6% | Claude Fable 5 |
| SWE-bench Pro | 80% | Coming soon | Coming soon |
| FrontierSWE v2 | 48.0% | Coming soon | Coming soon |
| FrontierCode 1.1 Main | 53.5% | 24.3% | Claude Fable 5 |
| Terminal-Bench 2.0 | 84.3% | Coming soon | Coming soon |
| cursorBench31 | 70.6% | 48.8% | Claude Fable 5 |
| cursorBench32 | 70.5% | Coming soon | Coming soon |
| VulcanBench v3 | 89.5% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 89.8% | 82.1% | Claude Fable 5 |
| SWE-bench (Vals) | 95.0% | 77.4% | Claude Fable 5 |
| SWE-Rebench | Coming soon | 60.7% | Coming soon |
| React Native Evals | Coming soon | 80.6% | Coming soon |
| Vibe Code Bench | Coming soon | 51.48% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- Claude Fable 5 public-lane score: 62.5 (Unranked · 2 rankable rows)
- Claude Sonnet 4.6 public-lane score: 54.1 (#33/48)

| Benchmark | Claude Fable 5 | Claude Sonnet 4.6 | Winner |
|-----------|-----------|-----------|--------|
| Blueprint-Bench 2 | 38.6% | Coming soon | Coming soon |
| OfficeQA Pro | 57.9% | Coming soon | Coming soon |
| CharXiv | Coming soon | 77.4% | Coming soon |

### Reasoning

- Winner: Coming soon
- Claude Fable 5 public-lane score: 76.2 (#11/22)
- Claude Sonnet 4.6 public-lane score: 65.8 (Unranked · 2 rankable rows)

### Knowledge

- Winner: Claude Fable 5
- Claude Fable 5 public-lane score: 83.5 (Supported · #2/181)
- Claude Sonnet 4.6 public-lane score: 56.7 (Supported · #51/181)

| Benchmark | Claude Fable 5 | Claude Sonnet 4.6 | Winner |
|-----------|-----------|-----------|--------|
| GPQA Diamond (Vals) | 93.2% | 85.6% | Claude Fable 5 |
| MMLU-Pro (Vals) | 91.5% | 87.3% | Claude Fable 5 |
| GPQA | Coming soon | 89.9% | Coming soon |
| SuperGPQA | Coming soon | 95% | Coming soon |
| MMLU-Pro | Coming soon | 79.2% | Coming soon |
| HLE | Coming soon | 49% | Coming soon |

### Instruction Following

- Winner: Coming soon
- Claude Fable 5 public-lane score: 78.3 (#54/120)
- Claude Sonnet 4.6 public-lane score: 47.9 (#84/120)

### Mathematics

- Winner: Coming soon
- Claude Fable 5 public-lane score: Coming soon
- Claude Sonnet 4.6 public-lane score: 49 (Unranked · 2 rankable rows)

| Benchmark | Claude Fable 5 | Claude Sonnet 4.6 | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | Coming soon | 32.400% | Coming soon |
| FrontierMath v2 (Tier 4) | Coming soon | 8.300% | Coming soon |

## FAQ

### Which is better overall, Claude Fable 5 or Claude Sonnet 4.6?

Claude Fable 5 is ahead overall on BenchLM right now.

### Where is the biggest gap between Claude Fable 5 and Claude Sonnet 4.6?

The widest category gap is in instruction following, where the averages are 78.3 for Claude Fable 5 and 47.9 for Claude Sonnet 4.6.

### How many benchmarks does Claude Fable 5 cover on BenchLM?

Claude Fable 5 currently has 18 sourced benchmark scores on BenchLM.

### How many benchmarks does Claude Sonnet 4.6 cover on BenchLM?

Claude Sonnet 4.6 currently has 25 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Claude Fable 5 vs Claude Fable 5.1](/compare/claude-fable-vs-claude-fable-5-1)
- [Claude Sonnet 4.6 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-claude-sonnet-4-6)
- [Claude Fable 5 vs GPT-6 Astra](/compare/claude-fable-vs-gpt-6-astra)
- [Claude Sonnet 4.6 vs GPT-6 Astra](/compare/claude-sonnet-4-6-vs-gpt-6-astra)

## Explore More

- [Claude Fable 5 profile](/models/claude-fable)
- [Claude Sonnet 4.6 profile](/models/claude-sonnet-4-6)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)