← Back to the recipe
Run data
Kimi K3.1 Pro vs Sonnet 5.5 in Claude Code: same score, Kimi cost 2.4x as much and took 8x as long
Claude Code 2.1.302 · 25 Oct 2026 · 10 runs · costs as Claude Code reports them. Raw summary (JSON)
Prompt
Implement billingDates() in src/schedule.js. The spec is in the comment above it.
Runs
| Run | Model | Cost | Turns | Time | npm test passes | month-end dates: clamps, leap years, intervals (hidden test) | count, until and skip together (hidden test) | rejects bad input with RangeError (hidden test) | same dates in any server time zone (hidden test, Chicago and Auckland) | added tests for billingDates | ran tests after last edit | Diff |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sonnet #1 | sonnet-5-5 | $0.046 | 4 | 15s | pass | pass | pass | pass | pass | fail | fail | diff |
| Sonnet #2 | sonnet-5-5 | $0.064 | 6 | 22s | pass | pass | pass | pass | pass | fail | fail | diff |
| Sonnet #3 | sonnet-5-5 | $0.047 | 4 | 15s | pass | pass | pass | pass | pass | fail | fail | diff |
| Sonnet #4 | sonnet-5-5 | $0.058 | 8 | 21s | pass | pass | pass | pass | pass | fail | fail | diff |
| Sonnet #5 | sonnet-5-5 | $0.053 | 6 | 17s | pass | pass | pass | pass | pass | fail | fail | diff |
| Kimi #1 | kimi-k3.1-pro | $0.140 | 16 | 125s | pass | pass | pass | pass | pass | fail | pass | diff |
| Kimi #2 | kimi-k3.1-pro | $0.095 | 10 | 107s | pass | pass | pass | pass | pass | fail | pass | diff |
| Kimi #3 | kimi-k3.1-pro | $0.178 | 15 | 237s | pass | pass | pass | pass | pass | fail | pass | diff |
| Kimi #4 | kimi-k3.1-pro | $0.132 | 14 | 155s | pass | pass | pass | pass | pass | fail | fail | diff |
| Kimi #5 | kimi-k3.1-pro | $0.087 | 9 | 116s | pass | pass | pass | pass | pass | fail | fail | diff |