Claude Fable 5.2 vs Kimi K3
Claude Fable 5.2 leads Kimi K3 by 21.5 points on the Shogo Agent Index. Claude Fable 5.2 wins 19 and Kimi K3 wins 0 of 19 shared benchmarks. Kimi K3 is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Claude Fable 5.2 | Kimi K3 |
|---|---|---|
| Agentic coding | 86 | 63 |
| Long-horizon autonomy | 78 | 65 |
| Knowledge-work agents | 67 | 42 |
| Computer use | 77 | 43 |
| Security & context learning | 65 | 53 |
Shared benchmarks
| Benchmark | Claude Fable 5.2 | Kimi K3 |
|---|---|---|
| APEX-Agentsindex | 80.1% | 47.8% |
| Vending-Bench 2index | $9,628 | $5,165 |
| FrontierSWEindex | 69.9% | 25.9% |
| PostTrainBenchindex | 44.6% | 31.5% |
| ARC-AGI | 98.5% | 94.5% |
| ARC-AGI-2 | 93.8% | 60.4% |
| Chess Puzzles | 79.0% | 37.0% |
| CritPt | 35.7% | 23.4% |
| DTBench | 99.2% | 91.2% |
| FrontierMath Tier 4 | 100.0% | 39.0% |
| FrontierMath Tiers 1-3 | 95.1% | 72.2% |
| GPQA Diamond | 96.0% | 93.1% |
| LMCA | 70.6% | 52.7% |
| OTIS Mock AIME | 99.7% | 97.2% |
| ProofBench | 100.0% | 87.0% |
| SciCode | 67.7% | 59.5% |
| SimpleBench | 94.9% | 60.7% |
| SimpleQA Verified | 94.0% | 50.6% |
| WeirdML v3 | 43.1% | 7.5% |
Specs and price
| Claude Fable 5.2 | Kimi K3 | |
|---|---|---|
| Agent Index | 76.4 (#1) | 54.9 (#26) |
| Agent $/M | $3.21 | $1.39 |
| Input / output $/M | $8.00 / $40.00 | $3.00 / $15.00 |
| Context | 1M | 1M |
| Released | Nov 9, 2026 | Jul 16, 2026 |
| Weights | Closed | Open |
| API hosts | 4 | 29 |
Don’t pick one. Route.
Shogo can run Claude Fable 5.2 and Kimi K3 side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo