# Claude Fable 5.1 vs Kimi K2.5

> Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 48.1 on the Noometry Index. Kimi K2.5 costs 22× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-fable-5-1-vs-kimi-k2-5
- Last updated: 2026-10-12
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. Claude Fable 5.1 scores higher in 9 categories and Kimi K2.5 in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Fable 5.1 leads 76.7 to 31.2.
- The biggest single-benchmark swing is ARC-AGI-2: 90% for Claude Fable 5.1 and 11.8% for Kimi K2.5.
- Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $10 / $50 for Claude Fable 5.1.
- Claude Fable 5.1 accepts more context: 1M tokens versus 262K.
- Kimi K2.5 has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 69.0 | 48.1 |
| Rank | 2 | 57 |
| Context | 1M | 262K |
| Input $/M | $10 | $0.45 |
| Output $/M | $50 | $2.25 |
| Weights | Proprietary | Open |

## Coding

- Claude Fable 5.1: 74.7 (#1)
- Kimi K2.5: 48.8 (#53)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| LMArena WebDev | 1744 | 1437 |
| SciCode | 63.1% | 49% |
| WeirdML | 92.9% | 45.6% |
| LMArena Coding | 1528 | 1474 |
| ALE-Bench | 2,143 | 821.65 |
| SWE-bench Verified | — | 73.8% |
| FrontierCode | 50.9% | — |
| SWE-bench Verified (bash only) | — | 70.8% |
| CursorBench | 51.8% | — |
| SWE-bench Multilingual | — | 67.3% |
| FrontierSWE | 56.3% | — |
| GSO | 88.2% | — |
| MirrorCode | 73.3% | — |

## Agentic & Tool Use

- Claude Fable 5.1: 50.7 (#5)
- Kimi K2.5: 34.2 (#48)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| Vending-Bench 2 | 5,422 | 1,198 |
| Terminal-Bench | — | 43.2% |
| APEX-Agents | 68.6% | — |
| Remote Labor Index | 17.9% | — |
| OSWorld | — | 63.3% |
| GDP.pdf | 29.6% | — |

## Reasoning

- Claude Fable 5.1: 76.7 (#7)
- Kimi K2.5: 31.2 (#80)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| ARC-AGI-2 | 90% | 11.8% |
| NYT Connections (extended) | 90% | 69.9% |
| ARC-AGI-1 | 97.5% | 65.3% |
| CritPt | 31.1% | 3.1% |
| Chess Puzzles | 47% | 12% |
| LMArena Hard Prompts | 1526 | 1453 |
| Epoch Capabilities Index | 164.7 | 148.03 |
| SimpleBench | — | 46.8% |
| Kagi LLM Benchmark | — | 78.5% |
| EnigmaEval | — | 3.4% |
| Thematic Generalization | — | 69.4% |
| EBR-Bench | 57.1% | — |
| Mystery Game Puzzles | 58% | — |
| DTBench | 97.6% | — |
| LMCA | 65.5% | — |

## Math

- Claude Fable 5.1: 89.6 (#4)
- Kimi K2.5: 51.8 (#53)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 92.2% |
| LMArena Math | 1525 | 1470 |
| FrontierMath (Tiers 1-3) | 90.2% | — |
| FrontierMath Tier 4 | 87.8% | — |
| MathArena Final-Answer Competitions | — | 62.3% |
| ProofBench | 100% | — |
| FrontierMath (Feb 2025 set) | — | 27.9% |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | — | 4.2% |

## Knowledge

- Claude Fable 5.1: 69.6 (#6)
- Kimi K2.5: 53.6 (#56)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| Humanity's Last Exam | 46.5% | 24.4% |
| SimpleQA Verified | 70.8% | 34.3% |
| LMArena Expert | 1535 | 1466 |
| GPQA Diamond | — | 87.6% |
| Vectara Hallucination Rate | — | 14.2% |

## Multimodal

- Claude Fable 5.1: 53.9 (#4)
- Kimi K2.5: 41.1 (#39)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| LMArena Vision | 1318 | 1269 |
| LMArena Document | 1513 | 1430 |
| Blueprint-Bench 2 | 41.9% | — |
| Furniture Assembly | 70% | — |

## Multilingual

- Claude Fable 5.1: 59.1 (#3)
- Kimi K2.5: 53.9 (#53)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | 1507 | 1433 |
| LMArena Chinese | 1586 | 1495 |
| LMArena French | 1525 | 1454 |
| LMArena German | 1500 | 1441 |
| LMArena Japanese | 1543 | 1421 |
| LMArena Korean | 1534 | 1410 |
| LMArena Russian | 1521 | 1435 |
| LMArena Spanish | 1516 | 1450 |

## Instruction Following

- Claude Fable 5.1: 79.2 (#6)
- Kimi K2.5: 75.3 (#64)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | 1517 | 1431 |

## Long Context

- Claude Fable 5.1: 46.7 (#20)
- Kimi K2.5: 52.1 (#7)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| LMArena Longer Query | 1522 | 1445 |
| Fiction.LiveBench | — | 86.1% |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |

## Writing & Preference

- Claude Fable 5.1: 79.2 (#2)
- Kimi K2.5: 65.1 (#53)

| Benchmark | Claude Fable 5.1 | Kimi K2.5 |
|---|---|---|
| LMArena Text | 1510 | 1445 |
| LMArena Creative Writing | 1507 | 1423 |
| EQ-Bench Creative Writing | 2162 | 1579 |
| LMArena Multi-Turn | 1492 | 1444 |

## FAQ

### Is Claude Fable 5.1 better than Kimi K2.5?

Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 48.1 on the Noometry Index. Kimi K2.5 costs 22× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.

### Which is cheaper, Claude Fable 5.1 or Kimi K2.5?

Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; Claude Fable 5.1 lists at $10 and $50.

### Is Claude Fable 5.1 or Kimi K2.5 better for coding?

Claude Fable 5.1 scores higher on coding benchmarks: 74.7 versus 48.8 in the Noometry coding category.

### Which has the bigger context window?

Claude Fable 5.1 does, with 1M tokens against 262K.

### How many benchmarks do Claude Fable 5.1 and Kimi K2.5 share?

34 benchmarks have published results for both models. Claude Fable 5.1 has 52 scored results on Noometry and Kimi K2.5 has 51.