Model comparison
Claude Fable 5.1 vs Kimi K3
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 59.5 on the Noometry Index. Kimi K3 costs 3.3× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.
Last verified . 43 shared benchmarks.
Summary
- They share 43 benchmarks with published results for both. Claude Fable 5.1 scores higher in 10 categories and Kimi K3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where Claude Fable 5.1 leads 53.9 to 37.8.
- The biggest single-benchmark swing is FrontierMath Tier 4: 87.8% for Claude Fable 5.1 and 39% for Kimi K3.
- Kimi K3 is cheaper at $3 / $15 per million input/output tokens, against $10 / $50 for Claude Fable 5.1.
- Kimi K3 accepts more context: 1.05M tokens versus 1M.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Claude Fable 5.1 | Kimi K3 | |
|---|---|---|
| Provider | Anthropic | Moonshot AI |
| Noometry Index | 69.0 | 59.5 |
| Released | 2026-09-01 | 2026-07-16 |
| Weights | Proprietary | Open |
| Context window | 1M | 1.05M |
| Max output | 128K | 1.05M |
| Input $ / M tokens | $10 | $3 |
| Output $ / M tokens | $50 | $15 |
| Results tracked | 52 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Fable 5.1 leads
Claude Fable 5.1: 74.7 (#1), Kimi K3: 61.0 (#10)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| FrontierCode | 50.9% | 44.2% |
| LMArena WebDev | 1744 | 1654 |
| FrontierSWE | 56.3% | 25.9% |
| SciCode | 63.1% | 59.5% |
| WeirdML | 92.9% | 82.6% |
| LMArena Coding | 1528 | 1508 |
| ALE-Bench | 2,143 | 1,524 |
| DeepSWE | — | 68.5% |
| CursorBench | 51.8% | — |
| GSO | 88.2% | — |
| MirrorCode | 73.3% | — |
Agentic & Tool Use Claude Fable 5.1 leads
Claude Fable 5.1: 50.7 (#5), Kimi K3: 41.8 (#20)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| APEX-Agents | 68.6% | 50.6% |
| GDP.pdf | 29.6% | 19% |
| Vending-Bench 2 | 5,422 | 5,165 |
| Remote Labor Index | 17.9% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
Reasoning Claude Fable 5.1 leads
Claude Fable 5.1: 76.7 (#7), Kimi K3: 63.0 (#17)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 90% | 60.4% |
| NYT Connections (extended) | 90% | 93.6% |
| ARC-AGI-1 | 97.5% | 94.5% |
| CritPt | 31.1% | 23.4% |
| Chess Puzzles | 47% | 39% |
| LMArena Hard Prompts | 1526 | 1496 |
| Mystery Game Puzzles | 58% | 26% |
| DTBench | 97.6% | 91.2% |
| LMCA | 65.5% | 52.7% |
| Epoch Capabilities Index | 164.7 | 157.45 |
| SimpleBench | — | 60.7% |
| EBR-Bench | 57.1% | — |
| Surface Evolver Bench | — | 95% |
| ForecastBench | — | 61.1 |
Math Claude Fable 5.1 leads
Claude Fable 5.1: 89.6 (#4), Kimi K3: 74.2 (#16)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 90.2% | 72.2% |
| FrontierMath Tier 4 | 87.8% | 39% |
| OTIS Mock AIME 2024-2025 | 100% | 97.2% |
| ProofBench | 100% | 87% |
| LMArena Math | 1525 | 1491 |
| MathArena Final-Answer Competitions | — | 87.8% |
| FrontierMath Erdős | 0% | — |
Knowledge Claude Fable 5.1 leads
Claude Fable 5.1: 69.6 (#6), Kimi K3: 63.2 (#21)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| SimpleQA Verified | 70.8% | 50.6% |
| LMArena Expert | 1535 | 1521 |
| GPQA Diamond | — | 93.1% |
| Humanity's Last Exam | 46.5% | — |
Multimodal Claude Fable 5.1 leads
Claude Fable 5.1: 53.9 (#4), Kimi K3: 37.8 (#70)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | 41.9% | 29.5% |
| Furniture Assembly | 70% | 34.2% |
| LMArena Vision | 1318 | — |
| LMArena Document | 1513 | — |
Multilingual Claude Fable 5.1 leads
Claude Fable 5.1: 59.1 (#3), Kimi K3: 56.3 (#21)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1507 | 1466 |
| LMArena Chinese | 1586 | 1529 |
| LMArena French | 1525 | 1491 |
| LMArena German | 1500 | 1488 |
| LMArena Japanese | 1543 | 1487 |
| LMArena Korean | 1534 | 1458 |
| LMArena Russian | 1521 | 1482 |
| LMArena Spanish | 1516 | 1472 |
Instruction Following Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#6), Kimi K3: 77.7 (#14)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1517 | 1483 |
Long Context Too close to call
Claude Fable 5.1: 46.7 (#20), Kimi K3: 45.8 (#29)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1522 | 1494 |
Writing & Preference Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#2), Kimi K3: 76.6 (#4)
| Benchmark | Claude Fable 5.1 | Kimi K3 |
|---|---|---|
| LMArena Text | 1510 | 1476 |
| LMArena Creative Writing | 1507 | 1454 |
| EQ-Bench Creative Writing | 2162 | 2082 |
| LMArena Multi-Turn | 1492 | 1488 |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is Claude Fable 5.1 better than Kimi K3?
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 59.5 on the Noometry Index. Kimi K3 costs 3.3× 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 K3?
Kimi K3 is cheaper. It lists at $3 per million input tokens and $15 per million output tokens; Claude Fable 5.1 lists at $10 and $50.
Is Claude Fable 5.1 or Kimi K3 better for coding?
Claude Fable 5.1 scores higher on coding benchmarks: 74.7 versus 61.0 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 1M.
How many benchmarks do Claude Fable 5.1 and Kimi K3 share?
43 benchmarks have published results for both models. Claude Fable 5.1 has 52 scored results on Noometry and Kimi K3 has 53.